IR 05000321/1986022: Difference between revisions

From kanterella
Jump to navigation Jump to search
(StriderTol Bot insert)
 
(StriderTol Bot change)
 
Line 1: Line 1:
{{Adams
{{Adams
| number = ML20214K287
| number = ML20215K584
| issue date = 05/14/1987
| issue date = 09/26/1986
| title = Ack Receipt of 861117 Ltr Describing Steps Taken to Correct Violations Noted in Insp Repts 50-321/86-22 & 50-366/86-22. Util Request for Withdrawal of Violation a Denied.Violation D Withdrawn
| title = Insp Repts 50-321/86-22 & 50-366/86-22 on 860728-0801. Violations Noted:Failure to Perform Matl Verification & to Adequately Document & Verify Matl Used During Valve Replacement
| author name = Grace J
| author name = Stadler S, Wilson B
| author affiliation = NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION II)
| author affiliation = NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION II)
| addressee name = Oreilly J
| addressee name =  
| addressee affiliation = GEORGIA POWER CO.
| addressee affiliation =  
| docket = 05000321, 05000366
| docket = 05000321, 05000366
| license number =  
| license number =  
| contact person =  
| contact person =  
| document report number = NUDOCS 8705280395
| document report number = 50-321-86-22, 50-366-86-22, NUDOCS 8610280220
| document type = CORRESPONDENCE-LETTERS, NRC TO UTILITY, OUTGOING CORRESPONDENCE
| package number = ML20215K512
| page count = 8
| document type = INSPECTION REPORT, NRC-GENERATED, INSPECTION REPORT, UTILITY, TEXT-INSPECTION & AUDIT & I&E CIRCULARS
| page count = 32
}}
}}


Line 18: Line 19:


=Text=
=Text=
{{#Wiki_filter:.
{{#Wiki_filter:p ucg  UNITED ST ATES
     [ hht
/  o  NUCLEAR REGULATORY COMMISSION
[  n  REGION li g  ,, j  101 MARIETTA STRCiET. *I  '*  ATLANTA, GEORCI A 30323
%...+/ +
Report Nos.: 50-321/86-22 and 50-366/86-22 Licensee: Georgia Power Company P. O. Box 4545 Atlanta, GA 30302 Docket Nos.: 50-321 and 50-366  License Nos.: DPR-57 and NPF-5 Facility Name: Hatch 1 and 2 Inspection Conducted: July 28 - August 1, 1986
    '
Inspectors: M . .
i[M[/4 S.~ Stadler ~    Date Signed H. O. Christensen W. K. Poertner G. A. Schnebli C. F. Smit J.,B. Brady J. D. Smith Accompanying Personnel: ' V. L. Brownlee, Branch Chief Approved by: hh, w w    #/)68$
B. Wilson, Acting Chief, Operational Programs  Date Signed Division of Reactor Safety SUMMARY
, Scope: This routine, special announced inspection was conducted in the areas of maintenance programs, implementation, and corrective action Results: Four violations were identified.
 
;
8610280220 861017
,
PDR ADOCK 05000321 0  PDR I
, , _ . -  . _ , - . _ _ __
     , , _ _ , . , - . - _ _ . , - _, - . , . . . -~
 
REPORT DETAILS Persons Contacted Licensee Employees
*Harvey Nix, General Manager
* Tom Seitz, Manager Maintenance
* Lewis Summer, Manager of Operations
* Tom Greene, Deputy General Manager
*Zachary Wahab, Engineering Supervisor
* Roger W. Zavadoski, Manager, Health Physics / Chemistry
* Jimmy Wilkes, Manager, Special Projects
*D. F. Moore, Nuclear Training Coordinator
*0. M. Fraser, Acting Quality Assurance Site Manager
*B. K. McLeod, Manager Maintenance and Outage
*P. R. Bemis, Manager Engineering General Offices
*B. R. Phillips, Training Supervisor
* O. Porter, Assistant Project Engineer
*R. A. Glasby, Project Manager
*A. Vest, PWPS Manager, Onsite
*P. A. Robertt., Acting Independent Site Engineering Group Supervisor
*S. H. Chesnut, Nuclear General Engineer
*R. K. Moxley, Quality Assurance
*S. J. Bethay, Acting Regulatory Compliance Superintendent
*C. R. Goodman, Acting Regulatory Compliance Supervisor
*G. A. Goode, Acting Manager Engineering i *C. T. Moore, Manager Training Other licensee employees contacted included engineers, technicians, operators, mechanics, and office personne NRC Resident Inspector
*P. Holmes-Ray (SRI)
* Attended exit interview Exit Interview
:
The inspection scope and findings were summarized on August 1,1986, with those persons indicated in the paragraph above. The inspector described the areas inspected and discussed in detail the inspection finding No dissenting comments were received from the license The licensee did not identify as proprietary any of the materials provided to or reviewed by the inspectors during this inspectio _)
 
_
 
3. Licensee Action on Previous Enforcement Matters This subject was not addressed in the inspectio . Unresolved Items Unresolved items are matters about which more information is required to determine whether they are acceptable or may involve violations or deviation Unresolved items identified during this inspection are discussed in paragraph . Maintenance Training Program The mechanical, electrical, and instrumentation and control (I&C) training courses are the last three courses to be submitted for INP0 accreditation by the~ licensee. The other six courses, licensed operator, non-licensed operator, requalification, shift technical advisor (STA), radiological protection, and manager and technical staff, were already INP0 accredite The inspector reviewed the Self Evaluation Reports (SERs) associated with the three remaining training programs to be accreoited, and the status of development. This review indicated that the licensee has prepared in-depth for the accreditation of these courses. A full job task analysis (JTA) was performed for each of the three maintenance positions, and all of the supporting materials such as texts and lesson plans will be completed prior to submittal for accreditatio In addition, the licensee has solicited heavy participation (up to 90 percent) by personnel from each of the three maintenance groups in the development of the JTA, courses, and related training material Involvement by job incumbents in the development of performance based training not only increases the technical accuracy and relativity, but can substantially increase the course credibility and i acceptance of the training by the various group At the time of this inspection, the status of development of these last three courses to be accredited were as follows:
-
Instrument and Control Technician: 119 of 141 modules completed or 84.4 percent
-
Electrician: 58 of 97 modules completed or 59.8 percent
'
- Mechanic: 45 of 93 modules completed or 48.4 percent The licensee indicated that development was on an accelerated schedule and
,
that all three courses were scheduled to be completed and submitted to INP0 by September 30, 1986 In an effort to upgrade the proficiency and performance of the Maintenance Group, the licensee is not only requiring that new hires and transfer personnel attend these courses, but also that all job incumbents complete the trainin Completion of the full training course by incumbents is projected to take up to three years for some individuals, and early indications are that there may be a substantial attrition rate due to the
__  . _ . increased level of training expectatio To counteract this expected attrition rate, the licensee is offering financial incentives for successful completion of the trainin Evening courses in remedial mathe-matics are also being offered on a voluntary attendance basis for all maintenance personnel. As pointed out in their June 12, 1986, presentation to the Region, the licensee has also increased the acceptance level criteria for new maintenance personnel. The inspector reviewed this upgrade in acceptance standards and noted that several significant changes had been made. A previously utilized written aptitude test was replaced in June, 1986, with the more comprehensive and widely used MASS test. The MASS test is administered to all non-transfer applicants for the mechanic, electrician, or instrument and control technician position The licensee believes the new test to be more relevant to the positions and a better indicator of potential success, as well as being more difficul In addition, the licensee discontinued the differential grading of written aptitude examinations for minority versus majority candidates in December, 1985. All applicants for maintenance positions are now subject- to the same acceptance criteria. The licensee has also added a skills test for mechanic and electrician applicant This test was developed to assess the
" hands-on" skill levels and was validated utilizing experienced job incumbents. Each skills test  takes about three hours and requires the applicant to demonstrate three basic skills applicable to the mechanic and electrician position The instrument and control technician, which is a non-union position, requires no skills test for applicants due to added experience requirements. The licensee estimated that this skills test would screen an additional 50 percent of the applicants for the mechanic and electrician position To facilitate the training of all maintenance personnel, including new hires and incumbents, the licensee is attempting to establish a shift schedule which will allow regular rotation through trainin A twelve-hour shift schedule was apparently rejected by the union, and there was inadequate staffing to support a six-shift schedule. Resolution of this schedule problem will be a key to ensuring that maintenance training is implemented on a timely and continuing basis. To allow maintenance training and qualification to occur on all shifts the licensee has appointed a full time training coordinator and assigned training evaluators on each shift. The evaluators will conduct on-shift evaluations and qualification checkoffs on specific job tasks for maintenance personnel in training. Although there are long term plans for a computer qualification card system, individual task qualifications at the time of this inspection were being tracked on new qualification card With increased maintenance staffing and the training of all incumbents, maintenance personnel will be in many different stages of training and qualified only on certain job tasks for some time. Considera-tion should be given to expediting the implementation of the computer qualification syste With adequate access terminals, this would allow maintenance supervisors to quickly access whether an individual is qualified for a particular task, or all individuals qualified for the tas This system should also help ensure that unqualified individuals are not assigned to perform maintenance tasks.
 
!
. , - . . - . - - . . _ - . _
  ,, , . _ . . - -. -. ,,.
 
The inspector toured the hands-on training laboratories established in the
'
training building for the Maintenance, Electrical, and I&C groups. These facilities had been inspected in the past and determined to be exemplar In preparation for INP0 accreditation and the implementation of the new performance based training programs, these superior facilities were being upgraded even further. The electrical laboratory contained a number of plant specific work / training stations and mockups. An electrical breaker is set up with all terminal connectors contained in a separate box. Trainees can disassemble and work on the breaker or trouble shoot problems in the terminal box. Breaker logics are also mimicked and instructors can insert logic problems utilizing a set of toggle switche Once the trainee has completed trouble shooting and repairs, the breaker can be operated to determine success or failure. Other training stations reviewed by the inspectors included duplication of various electrical relays, and a large valve with a Limitorque operator for trouble shooting and rebuilding. The electrical training course also includes 40 hours of training in the inter-pretation and use of electrical drawings, procedures, vendors manuals and other documents and the related symbols, acronyms, and terminology. The I&C Laboratory also was equipped with a large number of installed work stations and mock-ups. Where generic training stations or panels were purchased from vendors, they were modified to be plant specifi All training staff personnel interviewed indicated that plant management had been extremely cooperative in providing various training aids, such as pumps, valves, and breakers, so that the labs could be well equipped and plant specific. The
! I&C lab also had the capability to establish complete control loops for pressure, level, or flow instrumentation for trouble shooting and trainin This laboratory was particularly impressive and should allow duplicating most any major I&C repair, trouble shooting, or surveillance testing require-ments. The I&C training staff has also recently begun utilizing the plant simulator to train I&C technicians on responses of control room instrumen-tation such as neutron and power level monitoring, the rod worth minimizer l rod sequence control system, rod block monitor, and reactor water level controls. This is a relatively unique approach and could prove to be very l beneficial training. The maintenance laboratory was also very well equipped with various types of pumps and valves available for maintenance and trouble
; shooting. One pump was connected to a water storage tank so that maintenance trainees could conduct pump curve analysis following repairs and maintenance to the pump. This capability provides positive visual feedback to the
! trainees on their performanc We mechanical lab also emphasizes the I administrative aspect of complet r.) maintenance assignments. Trainees are
'
required to fill out and proce.s work requests, to fill out material and tool requisitions, and to complete and document post maintenance functional testing results.
 
! In summary, the corporate and plant support of maintenance training appeared I
to be comprehensive and logically developed, and the training facilities l probably among the best available anywher If the programs are implemented l as designed, and the union and scheduling problems resolved, the level of proficiency among maintenance personnel, and ultimately maintenance performance, should definitely increase. Since these training courses had been implemented only two weeks prior to this inspection, the implementation and effectiveness will be assessed during subsequent inspections.
 
l
:
,
j
 
6. Procedures Upgrade Program The procedures upgrade program (PUP) has been in existence at Hatch for the past three years. The phase I upgrade which concluded late in 1985 was apparently much less than effective, and did not result in any significant upgrade of procedures. Late in 1985, the primary contractor associated with Phase I was terminated and a new procedures upgrade group was formed. This new PUP group consists of approximately 60 full time personnel devoted to the upgrade of Hatch procedures, and this revised effort is being desig-nated as Phase II. To help ensure the success of Phase II, the licensee has centralized the PUP effort and is providing for increased Georgia Power involvement in the process. The licensee's objectives in Phase II of PUP are basically to ensure that all Hatch procedures are revised to be technically accurate, user friendly, and clear in intent to every qualified procedure user. To ensure that these objectives are met in Phase II, the licensee has implemented what appears to be a very rigid quality control program for procedure revisions. Plant management has provided nine full time personnel to assist in the procedure development, four from Maintenance and five from Operations. These personnel should not only assist in ensuring the technical accuracy and usability of the procedures, but their involvement in the development should instill a higher level of user confidence in the final produc In addition to the use of these subject matter experts, the licensee has incorporated several other controls in the development stage to ensure technical accuracy. The minimum qualifications for procedure writers has been procedurally defined as requiring 2 to 3 years in the nuclear field, a good technical understanding of boiling water reactors (BWRs) and procedural discipline, and previous procedure writing experience. For procedures related to systems and equipment, the procedure writers are required to cross check the vendor's manuals to ensure that applicable recommendations and requirements are incorporate The systems and equip-ment are also " walked down" as required to ensure the technical accuracy of terminology and locations versus actual plant conditions. The draft procedures are also verified against approved plant drawings to ensure accuracy. Finally the commitment tracking system is checked to ensure that all required commitments are incorporate The draft procedure review process is also tightly controlled and i procedurally defined. The initial reviews are performed internally by the technical advisors and by at least one other procedure writer. The draft procedures also undergo an as low as reasonably achievable (ALARA) review by Health Physics, and review and approval by Regulatory Compliance, QA and QC, Procedure Review Group (PRG), and the Plant Review Board (PRB). A new validation process which was incorporated into the procedure revision approval process approximately tWo months ago should add significantly to the technical accuracy, as well as the procedure credibility among users.
 
,
This validation process requires members of the user groups, other than those involved in the development process such as Maintenance or Operations
!
l l
 
personnel, to validate the new procedure by actual or simulated use. The methods of validation utilized, listed in the order of preference, include the following:
- actual performance of the procedure
-
use on the plant simulator (Operations procedures)
-
use on plant specific training mockups such as pumps and control stations
- walkthrough of the procedure without actually performing any work or evaluations
- in office comparison to the original procedure In response to inspectors concerns, the licensee indicated that the actual performance would always be the primary method of procedure validation subject to system or equipment availability and plant conditions. In addition, the in-office reviews against the original procedures would be limited to administrative type procedures, not operating, maintenance, or surveillance procedures.
 
The licensee intends to revise a total of approximately 5,000 procedures and had accelerated the schedule about one month before this inspectio All mechanical and electrical procedures (Maintenance Department) are scheduled to be completed within the next 6 months. The Instrument and Control (ISC)
procedure revision is not scheduled to be started until after the mechanical and electrical procedures are completed. The licensee based this schedule on the assumption that the I&C procedures were in better condition, requiring less priority. The inspectors noted that there were a number of I&C personnel errors in the last two years, and the events should be analyzed to ensure inadequate I&C procedures were not at faul t. If procedures were a large contributor to the error rate, this priority should be reevaluated. The general overall status of the PUP effort at the time of this inspection wasas follows:
-
840 drafts completed in 1986 of which 376 were in the maintenance area
- 46 percent of mechanical and electrical procedures completed
- 21 percent of all procedures completed
- 15 percent of I&C procedures completed
- only 4 procedures had been validated In sunnary, Phase II of PUP appeared at this stage of completion, to be an effective, ambitious, and well controlled effor Support by plant and corporate management was very good, including manpower for development and review. This manpower connitment needs to be extended to the validation process to ensure the continued quality of the procedures (only 4 have been validated to date). The confidence level in, and acceptance among procedure users of the new procedures appeared to be very high. Some confusion seemed to exist,'however, in how to determine a new procedure from the old in that the numbering system is identical. This confusion could cause employees to attribute inadequacies in an old procedure to a new procedure, and reduce
 
. -- -  -  .-_
 
:    7 confidence in the upgrade program. If this problem can be addressed, and the present level of development and review not downgraded to expedite development, the Hatch Procedures Upgrade Program could prove to be one of the more successful efforts of this typ . Maintenance Program (62702)
The inspectors reviewed the following procedures:
50AC-MNT-001-05: Maintenance Program i 50AC-MNT-007-0S: Preventive Maintenance Program
! DI-REG-10-0286N: Nuclear Plant Reliability Data System 51GM-MNT-007-0S: Control of Lubricants      6 30AC-0PS-005-0S: Temporary Bypass, Jumper, and Lifted Lead Control DI-MNT-02-1085: Maintenance History and Trending Program DI-REG-08-1285N: DR and LER Trending Program
 
51GM-MNT-002-0S: Maintenance Housekeeping and Tool Control i
j 40AC-ENG03-0: Design Control 40AC-Eng-009-0S: Control of Special Processes
.
.
May 14, 1987 orgia Power Company TTN: Mr. James P. O'Reilly Senior Vice President-Nuclear Operations P. O. Box 4545 Atlanta, GA 30302 Gentlemen:
30AC-0PS-002-0S: Plant Housekeeping and Cleanliness Control i
SUBJECT: REPORT NOS. 50-321/86-22 AND 50-366/86-22 Thank you for your response of November 17, 1986, to our Notice of Violation issued on October 17, 1986, concerning activities conducted at your Hatch facilit We have evaluated your response and found that it meets the requirements of 10 CFR 2.201. We will examine the implementation of your actions to correct Violations A, B, and C during future inspection After careful consideration of your request that Violation A be withdrawn, we have concluded, for the reasons presented in the enclosure to this letter, that the violations occurred as stated in the Notice of Violatio As described in the enclosure, the NRC has decided to downgrade Violation C from Severity Level IV to V due to limited safety significance. Additionally, Violation D is considered licensee identified and will be withdraw Should you have any questions concerning this letter, please contact u
'
30AC-0PS-001-0S: Equipment Clearance and Tagging 50AC-MNT-002-0S: Control of Measuring and Test Equipment 45QC-QCX-002: Quality Control Inspection Plans 53PM-MON-001-0S: Vibration Analysis
'
53PM-MON-002-0S: Lube oil Analysis I
42EN-INS-001-0S: ISI Pump and Valve Operability Program 40AC-FPX-001-05: Fire Protection Program The inspectors conducted a review of the above procedures to ascertain that the licensee has implemented a maintenance program that is in conformance with Technical Specifications, regulatory requirements, commitments, and
'
industry guides or standards. The inspectors also reviewed selected maintenance work orders (MW0), preventive maintenance procedures, conducted i
i
          , . ~ - -
  -_ _, . . , _ , . . _ . - . .
    - - _ . - - - - . . _ - - . _ _ . . , , . , . _ - . . _- , . , . , . ~ . - _


Sincerely, (Original signed by A. F. Gibson for)
i interviews with various maintenance personnel, and conducted tours of the plant to evaluate housekeeping practice The following aspects of the maintenance program were verified:
J. Nelson Grace Regional Administrator Enclosure:
-
Staff Assessment of Licensee Response cc w/ enc 1: (See page 2)
Written procedures were established for initiating requests for routine
! and emergency maintenanc Criteria and responsibilities for development, review and approval of maintenance requests were establishe Criteria and responsibilities that form the basis for designating the activity as safety or non-safety-related were establishe Criteria and responsibilities were designated for performing work  ,
inspection of maintenance activitie t
- Administrative controls for special processes were establishe Methods and responsibilities for equipment control were clearly defined and establishe Written procedures were established and responsibilities designated for cleanliness control of safety-related components and system ; - Administrative controls and responsibilities for general housekeeping were establishe Methods and responsibilities have been designated for performing functional testing of structures, systems or components following maintenanc Maintenance planning and scheduling is controlled by the Work Planning Group (WPG). The WPG is comprised of representatives from the Maintenance Department, Operations Department, Engineering Department, the Quality
. Control Section, and the Health Physics and Chemistry Department. The
!
responsibilities of the WPG as defined in 50 AC-MNT-001 are: Planning, scheduling, and tracking maintenance activities Generating work order packages Reviewing the work order packages l Generating and distributing schedules of maintenance Providing MWO data for entry into Nuclear Plant Reliability Data System (NPRDS) and other management information systems i Reporting the status of maintenance and modification activities on a periodic basis l
!    _ _ - -
_ _ _ - _ _ _ _ - . - _ . . _ . - - - _ . _ .
  - . - _


8705280395 870514 PDR ADOCK 05000321 G PDR
  . .- - - .
      \
9 Ensuring that required functional tests and inspections are adequate Ensuring that work descriptions and other documentation are adequate The WPG schedules work on a weekly basi Schedules are distributed to
Z QI
    ~
responsible supervisors on the Friday preceding the scheduled work wee All maintenance activities are tracked through a computerized system known as the Nuclear Plant Management Information System (NPMIS). All the information contained on an MWO is entered into the NPMIS. The licensee has instituted a program to generate MW0s for repetitive tasks using the NPMI This will allow the WPG to assume the function of scheduling the preventive maintenance program through the NPMIS. Previously preventive maintenance was scheduled by Regulatory Compliance, but at the time of the inspection,
' the WPG was using both NPMIS and the Regulatory Compliance generated pre-ventive maintenance schedules to track the preventive maintenance progra The inspector conducted a review of the licensee's implemented Preventive and Predictive Maintenance Program The licensee is presently in a
, Preventive Maintenance Procedure Upgrade Progra As of this inspection, the licensee had completed the review of all systems safety-related or important to operation, and determined the PM requirements for each compo-nent in the system. The licensee is presently in the process of verifying or incorporating these requirements into the PM procedures. Discussions with the licensee determined that approximately 30 percent of the PM procedures had been completed. During the review of the licensee's PM program, the inspector identified 34 PMs that were overdue as of July 28, 1986. A large portion of these PMs were overdue due to inaccessibility of the areas or to the plant conditions that presently exist. These overdue PM's are presently being tracked by the WP The licensee has instituted a program of predictive maintenance. This program consists of vibration analysis for pumps and turbines, oil analysis and infrared analysis. This data is used to trend the performance of each applicable component and is intended to reduce premature equipment failure and minimize unscheduled downtim The program is expanding in the number of equipment under evaluatio However, not all critical equipment is included, as evidenced by the recent diesel generator bearing failure. Some equipment requires some small modifications in order to draw representative oil samples. The diesel generator bearing is an example, and appears to be focusing management attention on the value of these programs and the necessity of including all critical equipment. The Predictive Maintenance Program is also used to defer performance of scheduled PMs based on satis-factory vibration analysis results and oil analysis results. Inspection Report 85-19 identified an inspector concern that the licensee had no administrative procedure that defined the Preventive Maintenance or Predic-tive Maintenance Program The inspector reviewed the administrative procedures associated with the Preventive and Predictive Maintenance
! Programs. Based on this review, the procedures do not appear to define the
;
Preventive and Predictive Maintenance Programs in sufficient depth.


._ _ _ ___ _ _ -
    -
  .
  .
Georgia Power Company  2  May 14, 1987 f.c w/ encl:
VJ. T. Beckham, Vice President,
  / Nuclear Operation VH. C. Nix, Site Operations General Manager
  /A. Fraser, Acting Site QA Supervisor
  ,/ L. Gucwa, Manager, Nuclear
  .
Safety and Licensing bec w/ encl:
RC Resident Inspector Hugh S. Jordan, Executive Secretary-Document Control Desk State of Georgia l
l
l
    .
  . - -_
mmw -_
y :,y p" w., n,  -
        '


Ernst S/87 i  . ..
The purpose of The inspectors conducted plant tours of both Units 1 and 2.
    -_ _ - _ _ _ _ _  I


    .  ._ _ . _
these tours was to assess the licenesee's housekeeping practices and to observe maintenance in progres The cleanliness of both units appeared satisfactor The inspectors were unable to find any safety-related maintenance in progress during these plant tours. Observation of mainte-nance activities will be inspected during a followup inspection.
 
The inspectors reviewed the Unit 2 clearance index to verify that the index was being maintained in accordance with procedure 30AC-0PS-001-0S, Equipment Clearance and Tagging. The inspectors also reviewed the Unit 2 control room MWO lo This review identified that the high pressure coolant injection (HPCI) inboard isolation valve 2E41-F002 had been electrically backseated on July 26, 1986, to stop excessive packing leakage. Electrical backseating of motor operated valves is accomplished at Hatch by bypassing the open limit switch and then driving the valve disc on the backseat until Discussions the locked with Operations, rotor current of the motor is approache Maintenance and Engineering personnel determined that electrical backseating of certain containment isolation valves is performed on a routine basi These valves are:
HPCI steam supply inboard and outboard isolation valves Reactor Core Isolation Cooling (RCIC) steam supply inboard and outboard isolation valves Reactor Water Cleanup (RWCU) suction supply inboard and outboard isolation valves Recirculating pump discharge valves Review of maintenance history files determined the following:
2E51-F007, RCIC steam supply inboard isolation, has been electrically backseated at least 17 times since July 1984, the last time the valve was disassembled for maintenanc E51-F008, RCIC steam supply outboard isolation, has been electrically backseated at least 18 times since March 1984, the last time the valve was disassembled for maintenanc E41-F002, HPCI steam supply inboard isolation, has been electrically backseated at least 10 times since February 198 As a result of a violation issued in NRC Inspection Report Nos. 50-321, 50-366/85-34 concerning valve 1G31-F001 exceeding its maximum closure time of 30 seconds due to being electrically backseated, the licensee has required an engineering evaluation be performed as part of the The backseating licensee procedure, and that an MWO be generated to repair the valve. These actions has also taken actions to prevent recurring packing leak include instituting a program to replace the packing on these valves on a regular interval and also plans to live-load the packing packing leakag The to help pr backseating containment isolation valves is an acceptable practic licensee presently has a request for engineering assistance initiated to
 
  . . -. _ - - - . -  - . --- --__ - - -  .-- .
;
 
'
provide guidelines for electrically backseating motor operated valves. The inspectors reviewed the electrical backseating procedure 52GM-MEL-02 If
;  electrical backseating is determined to be acceptable, the inspectors do not believe that the procedure is adequately as written. As written, the
'
procedure requires opening the valve from the full closed position using the temporary bypass around the open limit switch. The inspectors believe that
.!  the valve should initially be opened using the control room switch and then
,  eased onto its backseat using the temporary jumper around the open limit j  switch. The procedure does not require that continuity and megger checks be performed and documente The procedure also does not require that an MWO be written to inspect the internals of the valve for damage at the earliest
!  opportunity. As a result of this inspection effort, the inspectors deter-
,
mined that the licensee tests the stroke time of valves from light to light i  as opposed to switch to light. ASME Section XI requires that valves be j  timed from initiation of the actuating signal to the end of the actuating
,  cycl The licensee indicated that relief had been requested from NRR to I  allow timing from light to light vice switch to light, and that stroke time
:  testing .had always been performed from light to light at plant Hatch. No i  documentation was provided by the licensee that documented relief from the i
requirements of ASME Section XI.
 
f  In a subsequent phone call to Georgia Power Licensing, the inspectors
. requested copies of any documentation that requested relief from ASME l  Section XI. The inspectors do not consider that light to light stroke time
  ,
  ,
testing of valves is acceptable since this method does not take into account i  the time it takes the motor operator to start, the hammer blow effect
!  associated with the operator, and the time it takes the valve to travel to
!
the limit switch positio Technical Specifications require that certain
valves meet finite stroke time requirements. The inspectors feel that the
'
method of stroke time testing employed at plant Hatch is inadequate. Plant
'
Hatch also has no method of ensuring that all electrically backseated valves i  are timed from switch actuation during routine surveillance testing i  conducted after the: valve has been electrically backseated. This item was
!  identified to the licensee and will be reinspected in a later inspectio Electrical backseating of valves and light to light stroke time testing will be identified as an unresolved item (321,366/86-22-01).
j Review of Maintenance Procedures and Practices for Motor Operated Valves t  (M0Vs)
j
! The inspectors performed a detailed review of the licensee's procedures for maintenance on the safety related M0Vs installed throughout the plant. The review consisted of ensuring the requirements or guidelines from the vendor technical manuals (Limitorque), IE Notices and l  Bulletins, and general industry practices were contained in the
,
licensee's procedures. The following procedures were reviewed:
i
  *
52CM-MME-005-05, Rev. O, Limitorque Valve Operator Models SMB-0 through SMB-4, Mechanical Maintenance i  *
52CM-MME-006-0S, Rev. 0, Limitorque Valve Operator SMB-000 and SMB-00, Mechanical Maintenance i-
- _ . , , _ _ _ _ _ _ _  __ _ _ _ _ , _ _ _ - _ ._.. _ _ _ _ _ , _ .__ _
52CM-MME-007-0S, Rev. 0, Limitorque Valve Operator Model SMB-5 and SMB-5T, Mechanical Maintenance
*
52CM-MME-008-0S, Rev. 0, Limitorque Valve Operator Model SB-00, Mechanical Maintenance 52CM-MME-009-OS, REv. 0, Limitorque Valve Operator Model SB-0 through SB-4, Mechanical Maintenance 52 CM-MME-010-05, Rev. O Limitorque Manual Unit Models H0BC through H7BC, Mechanical Maintenance 52CM-MME-016-0S, Limitorque Valve Operator Type SMC-03, Mechanical Maintenance 52PM-MNT-005-0S, Rev. 3, Limitorque Valve Operator Inspection 52 GM-MNT-016-0S, Rev. O, Limitorque Valve Operator Trouble-shooting 52GM-MEL-022-0S, Rev. 1, Limitorque Valve Operator Electrical Maintenance 42SP-060986-P2-1-05, Rev. O, Inspecting Limitorque MOV Operators for EQ Compliance The inspectors considered the licensee's procedures for MOVs, in general, to be very good. The procedures flowed in logical step by step manner and were normally easy to follow. Steps contained in the vendor manuals that lacked clarity were written with more detail in procedures to prevent errors during repair. In addition, the
,
procedures were reviewed to ensure the following EQ requirements were i met:
-
The proper grease is specified for use in the actuator and limit switch gear box. All procedures specified the use of Exxon Nebula EP-0 or EP-1 for use in the actuator gear box and Mobil 28 for use in the limit switch gear box. These greases are required for use inside containment only, however, respoasible licensee personnel
! stated that all Limitorque valves would be worked utilizing the procedures reviewed and thus, these particular greases would
' be used in any valve repaired regardless of location or safety significanc Procedures required the use of the proper internal wiring between i terminal blocks, torque switch and limit switch. Any wires found
!
that could not be positively identified were required to be replaced.
:
i
_ ._. .  ,
.
l
      ,
  .
  .
ENCLOSURE STAFF ASSESSMENT OF LICENSEE RESPONSE Restatement of Violation A L
 
10 CFR 50, Appendix B, Criterion V, and the licensee's accepted QA program (HNP-2, FSAR-17, Section 17.2.5) require that activities affecting quality I shall be prescribed by documented procedures of a type appropriate to the circumstances and shall be accomplished in accordance with these procedure Hatch procedure 50AC-MNT-001-05, Maintenance Program, requires that deficiencies or nonconforming equipment noted during maintenance activities shall be documented and controlled in accordance with 10AC-MGR-004-0,.
      ;
Deficiency Control System, and documented on the maintenance work order (MWO)
-
in " Actual Work Performed" (Block 27). If it is necessary to change the scope of work during maintenance activities, an MWO will be written to cover the chang Hatch Procedure 10AC-MGR-004-05 also requires that a Deficiency Report be written for deficiencies on nonconforming equipment identified during maintenance activitie Contrary to the above:
Procedures required the proper torque switches and limit switche '
- The HPCI turbine failed to meet acceptance criteria under maintenance work order (MW0) No.1-86-3998 on April 30, 1986, and the licensee failed to write a new MWO to cover an expansion in scope of the maintenance activitie A procedural deficiency was identified during the calibration of a turbidimeter under MWO 1-84-4872 on May 31, 1986, and the licensee failed to initiate a Deficiency Report to document the deficiency and ensure prompt corrective actio An investigation of the failure of a fuel pool cooling pump to start conducted under MWO 1-86-3426 on April 10, 1986, determined a discrepancy ,
Switches inside containment were required to be dark-brown or off-white while those located outside containment could be dark-brown, off-white, or re Missing T-drains or incorrectly installed T-drains on inside containment actuators are required to be installed correctly at the lowest point on the motor to ensure moisture does not short out the moto The following procedural deficiencies were discussed with responsible licensee engineers. Although there appear to be no specific regulatory requirements concerning these items, they felt the concerns were valid and will consider implementation in future revisions to the procedure The procedures did not
between actual and rated full load current. No Deficiency Report was written on corrective action take Summary of Licensee's Response The licensee argues that although the events occurred, there were no actual or potential safety consequences. The licensee stated that the involved plant personnel erred in not initiating the required MWO to cover changes in
-
52CM-MME-006-0S and 52CM-MME-008-0S:
contain specific instructions to identify that the tripper fingers on the clutch tripper assembly should not be the same lengt Replacement fingers obtained through the supply system may be the same length and must be shortened by grinding or filin GM-MEL-022-0S: The procedure should provide a more positive means to ensure the motor pinion gear is properly installed or
,' reinstalled. The position of the motor pinion gear is critica The pinion must be mounted with the gear hub facing the motor on SMB-0 actuators. On SMB-1 through 4 actuators, the hub must face opposite the motor. Installation in the incorrect position mayAs cause gear tooth failure due to insufficient tooth engagemen presently written, the procedure only requires the installed position of the pinion be note GM-MEL-022-0S: The procedure should contain instructions to I -
ensure the torque switch is balanced during installation.
 
l l
j
-
52CM-MME-007-0S, As previously stated the maintenance procedures were well written, however, a review of this procedure indicated it was basically a word for word copy from the vendor technical manua The other procedures contain numerous additions and j clarifying instructions not contained in the technical manuals.
 
l This is due to increased experience from valve actuator mainte-nance. The licensee stated the reason this procedure was lacking compared to the others is that there are very few SMB-5 and SMB-5T actuators installed in the plant. Thus, the experienced gained from working the large number of the other types of actuators was not availuble for these actuator types. The licensee stated that as more maintenance experience was gained it would be incorporated into the procedure, as applicabl ._ _ _ _ _ _ _
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _
a
 
      .
 
b. Discussions with licensee engineers and maintenance personnel, and a review of internal correspondence, indicated that Hatch has had several problems associated with Limitorque operators. Plant Hatch has approximately 650 Limitorque motor operated valves, of which 220 are safety related. The problems discussed included the following:
-
Inspections at Plant Hatch revealed deficient grease in the main gear housing, the limit switch gear box, and/or the spring pack in 28.6 percent of the actuators inspected to date. In addition, separated grease was found in the majority of the actuators inspected in the warehouse. These problems are due to mixing of greases, harsh environments, or the tendency of grease to separate over long periods of tim To resolve and prevent grease problems which could cause improper Limitorque operation, several steps have been taken at Plant Hatch:
  (1) Identification of actuators with mixed greases and disassembly, cleaning, and relubrication with Exxon Nubela EP- (2) A memo was sent to Maintenance personnel, the lubrication guide was changed, and procedures were changed to note the use of Exxon Nebula EP-0 or EP-1 in Limitorque actuator (3) Lubrication analysis procedure 50PM-MON-002-0S was writte In addition to analyzing the oil, this procedure may be used
;  to sample and analyze the condition of grease without disassembly of the Limitorque actuator. Use of this method
,
will provide more confidence in the condition of grease in i  Limitorque actuators.
 
l l (4) Change over to Mobil Grease #28 for use in limit-switch gearboxes for higher temperature resistanc Several Plant Hatch actuators have also been found without torque switch limiter plate However, in lieu of relying on limiter plates to ensure the torque switch setting does not exceed the maximum value recommended, new Limitorque corrective maintenance
.
procedures and the new set-up and test procedure will require l listings of "as found" and "as left" settings with notes to not exceed the maximum settin In addition, training procedures for setting Limitorque torque and limit switches have been evaluated to ensure corrective methods are use Of all the Limitorque actuators inspected at Plant Hatch, 2 l percent had deficient limit switches and/or torque switches (not
; including grease related problems). In 7 cases limit switch l
rotors and/or baseplates were found cracked or broke Replace-
-
ment of deficient switches is being done as they are identified
!
l l
l
!
;
- - _ _ . . _ _ _ , __ _ _ _ _ _ - _
 
in inspection New procurement policies and maintenance proce-dures will ensure that the correct switches (material type) are use Several operators have been found without the drilled drain plugs (T-drains) installed as per Limitorque instruction As a result, a memo will be sent to all Maintenance personnel concerning the installation of T-drains in Limitorque motors with
"RH" insulation. In addition, all actuators with RH insulated motors will be inspecte T-drains will be installed on any motors without the drains installed according to Limitorque's instruction All gaskets are replaced during reassembly of the operators after inspection. In only one case have the gaskets been found damage In one other case a gasket with the wrong thickness size was found as the cause of mechanical problems with the actuato Another problem has also been identified in Plant Hatch inspec-tions to date. Of the Limitorque operators inspected, 34.7 percent have been found to have problems with the declutch to manual operation " system". These problems were caused by poor operating knowledge, i.e., trying to force the declutch lever to put the operator in the manual mode. Bent or twisted declutch shafts were found in 22.5 percent of the operator To reduce these type problems, directives were sent to both Maintenance and Operations personnel instructing them on the proper methods of putting a Limitorque into " hand" operation. Also, training procedures have been evaluated to ensure proper instruction is give The licensee's program and procedures for maintenance, inspection, and problem resolution for M0Vs appeared to be adequate. It should be noted that implementation of the program and actual adherence to the procedures was not evaluated during the course of this inspection, due to lack of maintenance in progress. These items will be addressed in a future inspection when actual maintenance in progress can be observed.
 
9. Maintenance Implementation (62700)
The inspectors reviewed selected maintenance work order packages (MWO),
procedures, maintenance history files and interviewed several maintenance supervisor The maintenance work associated with Licensee Event Report (LER) 86-027, Valve Leakage Found During Local Leak Rate Test, Required Rector Shutdown, was reviewed. On June 12, 1986, Hatch Unit I was required to shut down due to the failure of drywell vent valves, IT48-F319 and IT48-F320, to pass the required local leak rate tes Maintenance work orders (MWO) 1-86-5083 and 1-86-5084 were written to repair 1T48-F-319 and IT48-F-320 respectivel The inspector noted that MW0 1-86-5083 and MW01-86-5084 indicated in block 27 of the work order continuation sheets that the valves 1T48-F-319 and
 
1T48-F-320 were repaired and replaced in the system on June 14, 1986. On June 15,1986, both valves were again removed from the system. The MW0s did not state why the valves were removed from the system, or why they were being reworked. The licensee informed the inspector that the valves were removed from the system for rework because they failed a preliminary leak rate test. This test was not documented in the MWO package. Administrative control procedure 50AC-MNT-001-0S, Maintenance Program, Step 8.6.2 requests that the results of all inspections and tests performed shall be documented and further action based on results and the documentation will become part of the MWO package. The licensee failed to document the results of the preliminary test associated with MW0s 1-86-5083 and 1-86-5084, concerning repairs on drywell vent valves 1T48-F319 and IT48-F320. The licensee was informed that the above failure to complete the MW0 per established proce-dures was a potential violation (321/86-22-02). Additionally, a review of MWO 1-86-5084 indicated the following deficiencies:
- New i inch copper tubing with new nuts and ferrels were installed on valve IT48-F319 operator with out a stock material issued (SMI) form being completed and Quality Control did not perform a material veriff-cation. Maintenance administrative procedure 50AC-MNT-001-05, revision 4, dated November 27, 1985, requires that if material is needed during MWO activities, an SMI should be completed in accordance with material procedures and Quality Control will perform material verification activities as part of the QC inspection program. The licensee was informed that this was another example of potential violation (321/86-22-02). Other examples of failure to adequately document and verify the material used includes MWO 1-86-0389, replacement of resilient seats in valve 1T48-F320, where Swageloc brand fittings and i inch copper tubing from the maintenance shop stock were used to make repairs to valve IT48-F320. The work description for MWO 1-84-8358, repair of valve 1T48-F319, states: removed operator and found one key was missing and the other key was half sheared, also sleeve where key fits was wallowed ou The replacement of the keys with proper
' material was not documented or QC verifie The torque wrench used to torque the flange bolt and operator mounting bolts for valve 1T48-F319 was not documented in block 27 (actual work performed section) of the MW0. Revision 5, dated July 29,1986, of maintenance procedure 50AC-MNT-001-05 requires documenting identifica-tion numbers of the calibration tools used, and the calibration due date of each tool in block 27. The quality control (QC) inspector did
,
,
work scope and that deficiency reports (DRs) were not initiated as required
document the tools used on the QC inspection checklist. MWO 1-86-5083, j
'
'
because personnel did not clearly understand when a DR was required or who was
for valve IT48-F320 also lacked calibration tool documentation in block 27 of the MW Other examples of where torque wrenches were not documented as being used are MWO 1-84-8403 and MWO 1-84-8358.
: responsible for initiating the DR because the DR procedure was not sufficiently J clear in those area _ _ _ ___ _ . _ _ . _ _ __ _
 
_ . _ _ _ _ _ - __ _ _ _ -
t i
The review of MWO 1-86-5083 and 1-86-5084 indicated that the valve flange bolts were torqued to 675 foot pounds using a 250 foot pound torque wrench l in combination with a torque wrench multiplier. A review of measuring and test equipment records showed that the torque wrench was controlled, calibrated and properly identifie The torque wrench multiplier had
 
identification number, 391A268517, but no documentation other than the operating instructions for Model 391A torque multiplier to identify the multiplier and its accuracie The Model 391A torque multiplier uses a planetary geared action to produce a 6 to 1 torque ratio with a stated accuracy of 5 percent. The multiplier has a rated input capacity of 200 ft-lbs with a rated output capacity of 1200 ft-lb CFR 50, Appendix B, Criterion X11, states: Measures shall be established to assure that tools, gages, instruments, and other measuring and testing devices used in activities affecting quality are properly controlled, calibrated, and adjusted at specified periods to maintain accuracy within-necessary limit The licensee failed to assure that the torque multiplier was properly calibrated or adjusted to maintain accuracy within necessary limits. The licensee informed the inspector that they contacted the manufacturer of the torque multiplier and the manufacturer stated that each torque multiplier is calibrated during assembly and its torque ratios and accuracy is uniquely determined and documented. The licensee stated that they did not receive this manufacturer's documentation or test the multiplier themselves to determine its accuracy. The licensee was informed that failure to ensure the torque multiplier's accuracy was a potential violation (321, 366/86-22-03).
 
Additionally, the use of a torque multiplier with a given torque wrench may cause the mechanic to under or over torque a bol The torque wrenchs are calibrated with an allowable accuracy of 4 percent of full scale in the clockwise direction and 6 percent of full scale in the counter clockwise direction. That would allow a 250 ft-lb torque wrench to have a 10 ft-lb error band for clockwise operation A 1000 ft-lb torque wrench would have a 140 ft-lb error band. If a mechanic torques a bolt to 675 ft-lb using a 100 ft-lb torque wrench, the bolt could be torqued to 675 ft-lb 40 ft-l If the mechanic used a 250ft-lb torque wrench in combination with a 6 to 1 torque multiplier, the bolt would be torqued to 675 ft-lb 60 ft-lb, due to the multiplication of the torque wrench (112.5 ft-lb) setting plus its maximum allowed accuracy range of 10 f t-l The accuracy of the torque multiplier is 5 percent, and if 100 ft-lb was applied, the results could be 600 30 f t-lb. The additional error of 30 ft-lb would also effect the total torque value on the bolt. The licensee informed the inspector that they would review the tolerances associated with bolt torque requirements and determine if the torque wrench accuracies in combination with torque multipliers is acceptable. The above determination will be an inspector followup item (321,366/86-22-04).
 
The inspector reviewed several MW0s associated with the calibration of the turbine building main steam leak detection instrumentation, IU61-N101-A through D. When performing the calibration on October 16, 1984, two instruments were found out of tolerance, IU61-N101A and B. On March 28, 1985, Unit 1 received an invalid half group I isolation on reactor protec-tion system (RPS) "A" channel due to IU61-N101A and 1U61-N101B being out of tolerance, which caused a RPS actuation signa When performing the latest calibration of these instruments, on December 4, 1985, all 1U61-N101-A through D temperature instruments were out of tolerance. The L
 
inspector expressed concern about the instrumentation trending, and the cause for the instrumentation drift. The instrument and control supervisors indicated that they believed the drifting was attributed to the environ-mental conditions of the turbine building when the instruments were calibrated, i.e., high temperature condition in the summer, and lower temperatures during the fall and winte The inspector requested the deficiency report or the request for engineering (RER) review on these instruments. The review of the deficiency report or RER will be an inspector followup item (321/86-22-05).
 
10. Corrective Action (92720)
' General The inspectors performed an inspection of corrective action in the maintenance area to assess the status of this functional area, and to determine whether the licensee had developed a comprehensive corrective action program to identify, follow, and correct safety related problems. The following criteria were used during the review to assess the overall acceptability of the established program:
  - Had the licensee established policies and administrative controls for identifying problem Had Management controls been established for the tracking and resolution of problems identified by:
Operational events Internally identified problems Quality Assurance Audits
>
NRC inspection findings Employee concern programs and concerns brought by external persons or organizations The documents listed below were reviewed to determine if these criteria had been incorporated into the corrective action program.
 
!  10AC-MGR-04-0, Deficiency Control System Revision 0 10AC-MGR-005-0S, Corrective Action Trending and Tracking, Revision 1 AG-MGR-02-1284, Quality Concern Program Revision 1 AG-REG-04-185N, NRC/QA Item Corrective Action Program, Revision 0 DI-REG-07-1085N, NRC/QA/INP0 Item Corrective Action Program, Revision 1 43RC-CPL-003-0S, Defects and Noncompliance 0
.-. _
_ _ _ _ _ _ _ _ . _ __ _ _ _ _ _ _ _ _ _ _ _ - _ _ _ _ _ .


.
QA-05-01, Field Audits, Revision 15 QA-05-02, Corporate / Supplier Audits, Revision 14 QA-05-13, NRC Open Items Control, Revision 13 QA-05-17, QA Surveillance, Revision 2 QA-05-20, QA Trend Program, Revision 2 DI-QCX-02-0485, Quality Feedback Report, Revision 0 During the inspectors' review of the corrective action programs, examples were found that raised concerns in the areas of:
Enclosure  2 The licensee indicated that Deficiency Report N was written on November 7, 1986, to address the cited instances and that control of work scope authorized by a particular MWO would be re-emphasized with the Maintenance Department in training meetings by December 10, 1986. The licensee states that prior to the NRC inspection, site Quality Assurance (QA) had identified inadequacies in the Deficiency Control System procedure (10AC-MGR-004-05). In addition to resolving the QA concerns, the licensee was revising the procedure to make it easier for plant personnel to understand when a DR must be initiated and by whom. Part of this corrective action would be the appropriate training of plant personnel on the procedure revision. The procedure was scheduled to be implemented by December 26, 198 The licensee requested that Violation "A" be downgraded to an Inspector Followup Item based on the fact that procedure 10AC-MGR-004-0S was in the process of revision at the time of the citation due to a licensee identified problem. The licensee contends that the corrective action for the licensee identified item would have prevented the cited violatio NRC Evaluation For a violation to be considered licensee identified, it must have been identified by the licensee; fit in a Severity Level IV or V category; be reported if required; and most importantly, it should not be a violation that
Failure of corrective actions to prevent reoccurrence The failure of event cause analysis to determine actual root cause Too narrow scope in identifying previous similar events in LER The description of events in some LERs did not clearly describe all aspects of the even Excessive personnel error rat Continuing and repetitive reactor water cleanup (RWCU) system isolations from high room temperature and high differential flow over a two-year period are an example of inadequate corrective actions. The majority of these engineered safeguards feature (ESF) equipment actuations were caused by high RWCU room temperatur Several high differential flow actuations, however, resulted during valving operations for the RWCU filter /demineralizers. These isolations appear to have been caused by operating the valves to fast, and a failure to follow procedures. The licensees root cause analysis and corrective actions appeared to be less than timely for correcting the RWCU system isolations caused by high room temperatures. A new ATTS system was installed in 1986 which lowered the room temperature trip setpoint from
'
145 F to 124 F, and physically mounted the new temperature sensors in the overhead of the room. The lower temperature setpoint established by this modification resulted in an increase in ESF actuations (approximately 12 actuations over a 3 month period). The licensee stated that the trip setpoint was lowered to this value because adequate data was not provided to General Electric (GE) for analysis prior to the installation. This data has subsequently been provided to
,
,
GE and their analysis will permit the room temperature setpoint to be raised to 150 F which should resolve this problem. The concern is that the measures taken during this 2 year period did not promptly correct the identified problem which resulted in excessive challenges to the ESF equipmen Another example of event analysis deficiencies was the
assumptions made without technical basis in LER 85-043, Failure of Containment Penetrations to Pass Local Leak Rate Tests. The statement was made in the analysis section of this LER that the degradation of the valves to the point that they could not pass the leakage tests u probably occurred toward the end of the surveillance interval. Subse-quent failure of valves identified in this LER required a plant shut-down to repair the excessive leakage. Licensee management stated during interviews that they eventually performed additional analysis for the valves identified in LER-043 to better support the analysis statement. A significant number of events reported in accordance with 10 CFR 50.72 and 50.73 that were reviewed had the cause of event listed as unknow Although the exact cause of events cannot always be identified, the proportionate number of unknowns appeared excessiv Additionally, a number of reportable events reviewed did not list previous related events as similar. An example were isolations of the RWCU system which occurred on December 21, 1985, and January 10, and 15, 1986. All of these events identified the cause as personnel error, two I&C technician errors and one system alignment error. Although each of these events involved inadvertent RWCU isolations by I&C personnel, and in two of the events improper use of jumpers, each report indicated no previous similar even It appears that the licensee classification of similar events may be too narrow in scope to ensure that trends are identified and corrective actions taken to prevent recurrenc The LERs and event reports reviewed, indicated a relatively significant amount of personnel errors are continuing to occur at Plant Hatch. Although the inspectors realize that totally eliminating personnel errors is virtually I impossible, many of the personnel errors reviewed indicated a potential for reduction or elimination. The corrective actions necessary to reduce the error rates might include many different directions including procedure revisions, procedural adherence, improvements in
'
'
could reasonably be expected to have been prevented by the licensee's corrective action for a previous violatio Contrary to the above requirements, the corrective actions stated in the licensee's response, i.e., training of maintenance personnel on control of work scope and the revision of procedure 10AC-MGR-004-0S did not occur until December 10 and 26, 1986, respectively. The three examples cited in the violation occurred in April and May 1986, and were not prevented by the j corrective actions implemented in Decembe In addition, a similar violation was cited on December 26, 1984, (50-321,366/84-46-02), and the corrective actions implemented appear neither timely nor adequate. The most recent event involving accidental release of spent fuel pool water appears to be a further indication of a generic failure of the corrective action program to ensure effective corrective action for deficiencies identified during maintenance activitie NRC Conclusion For the above reasons, the NRC staff denies the licensee's request to downgrade the violation to an Inspector Followup Ite The three examples cited in the ,
labeling and human factors areas, independent verification, additional training, and system design changes. Many of the new programs being i established by the licensee have the potential to meet these objectives and to significantly impact the personnel error rates, providing proper emphasis is placed on identifying all contributing causes and trending l similar events. At the time of the inspection, however, these programs were in the early stages of implementation and had not produced a notice-able effect on Hatch personnel error rates. This area will be reviewed again during the maintenance followup inspection to be conducted, b. Deficiency Reporting System The inspectors determined that licensee management had established various administrative control systems to identify and correct conditions adverse to quality. The administrative controls employed l for identifying and resolving defects are as follows:
violation occurred approximately six months prior to the implemented corrective I actions, and do not meet the requirements of Part 2, Appendix C, for licensee identified violation _ _ _
A maintenance work order (MW0) is used to control the correction of defects in station equipmen .
.


      -. - - -._ - _
A request for engineering review (RER) is used to initiate and control modifications to station equipmen The material deficiency section of a Deficiency Report (DR) is used to control and process non-conforming materials, parts, and components identified during receipt inspection A temporary procedure change form or a procedure / instruction request form is used to control the development of or revision to Plant Hatch procedures and instruction The training request or training directive is used to identify, request, and document the training needed when a problem was caused by personnel erro In addition to the administrative controls described above, a deficiency report (DR) is prepared for deficiencies which have been classified as significan This ensures that appropriate management review and root cause analysis are perforited for identified deficiencies. Guidance in the classification of significant conditions adverse to quality has been delineated in administrative procedure 10AC-MGR-04-0. Organizational responsibilities for ensuring effective implementation of the Deficiency Control System has also been delineated in writin c. Quality Trend Programs Licensee management has established a Corrective Action Trending and Tracking Program. This program ensures that problems identified under the deficiency reporting system are assessed for recurrence, in addition to providing a formal process for assessing and tracking problems identified by the NRC, INP0, the Safety Review Board (SRB) or Quality Assurance Organization. Administrative procedure 10AC-MGR-005-0S delineates the methods and assigns responsibilities for:
_ . - -- . -- .- _ - .
Trending material / operational problems to identify, correct, and report program, operational and generic material deficiencie Assessing and correcting deficiencies identified by outside agencies or corporate inspection and audit program Assessing and correcting generic industry problems identified throughout the nuclear industry and which are considered applicable to Plant Hatc .- -     __  _
  .
  .
,
  .
  .
Enclosure  3 Restatement of Violation 8 NRC Confirmatory Order 7590-01, dated July 10, 1981, required implementation of NUREG-0737 Item I.C.5, Establishment of Procedures for Feedback of Operating Experience to Plant Staff. Paragraph 7 of Item I.C.5 requires that the feedback program functions effectively at all levels.


i Hatch procedure 30-AC-0PS-003, Plant Operations, Section 4.2.3, requires that the Engineering Department conduct an audit of the functioning of the Operating i Experience Feedback Program at all levels on a biennial frequenc Contrary to the above, audits of the functioning of the Operating Experience Feedback Program at all levels have not been performed and documented by the Engineering Department.
The scope of the trend program includes monitoring deficiencies on any system, structure, or component whose failure was the cause of a condition to be reported by a licensee event report (LER), or by the reporting requirements of 10 CFR 2 Responsibility has been assigned to the Regulatory Compliance Department for monitoring DRs that identify non-conforming material / operational conditions which have been determined to be reportable in accordance with the requirements of administrative procedure 10 PC-MGR-04- Nonreportable DR's are also trended by Regulatory Complianc The Corrective Action Trending and Tracking Program provides for the performance of a monthly trend sort to evaluate potential trend Adverse trends are reported to the superintendent QC and the cognizant department manager or his designee. The program requires that a DR be prepared by the cognizant department manager to ensure evaluation and root cause analysis by plant managemen The inspectors reviewed the following documents in connection with the assessment of the corrective action trending and tracking program:
NRC/QA item trend monthly reports for the period June 1985 to January 198 NRC/QA item trend monthly reports for the period December 1985 to June 1985 The inspectors concluded that the Corrective Action Trending and Tracking Program appears adequate. Implementation of the program is less than fully effective however, based on the following observations:
*-
Increase in backlog items from 35% in May to 60% in June as described in report dated July 8, 198 .
A total of 14 items requiring manageaent's attention over the one f
year time span reviewed.


!  Missed milestone dates are attributed by the licensee as the cause for the above. The arrangement for Regulatory Compliance to provide
,  department managers with a " Weekly Overdue Item Status Report" will hopefully correct this lack of adherence to established milestone dates for prompt correction of identified deficiencie Regulatory Compliance has been assigned responsibility for the detailed implementation of the corrective action program through all phases of receiving, issuing, and processing NRC/INP0/QA items. The administra-tive control for implementation of the corrective action program l  delineated in administrative guideline AG-REG-04-1085N and department l
instruction DI-REG-07-1085N were reviewed by the inspectors and found
'
'
This is a Severity Level IV violation (Supplement II).
to meet regulatory requirements and licensee commitments.
 
i
. ,- -. - ,-. , , - - - - . - - . - - - . , . , ~ , , - , . . , . . - . - . - - _ . , , _ . _ - - _ _ - - _ _ - - , - . .
 
.


Summary of Licensee's Response The licensee noted that it is subsection 8.11.3.4 of 30AC-0PS-003-0S which requires the biennial audit of the Operating Experience Feedback Program. The licensee stated that due to personnel oversight, Engineering Department procedures or instructions were not prepared and approved to perfonn the
d. Quality Assurance Audits Technical Specification (TS) Section 6.5.2.8 delineates the require-ments for the conduct of audits under the cognizance of the Safety Review Board (SRB). The inspectors reviewed the following documents and verified that audits of corrective actions specified in TS section 6.5.2.8.2.c are conducted at the required frequency:
,- required audit. The licensee noted that the August 1986, INP0 evaluation identified (number OE.1-1) that the effectiveness of the existing Operating Experience Feedback Program was not periodically assesse A procedure was
Plant Hatch QA Department, Annual Audit Planning Matrix, Schedule and Status Report for the year 1985, dated 12/5/85 Plant Hatch QA Department, Annual Audit Planning Matrix, schedule and status for the year 1986, dated 5/13/8 The following audits were reviewed by the inspectors to verify compliance with program procedures and regulatory requirement Audit Number: 85-CA-1 Date Audited: March 18,1985 - April 11,1985 Activity: QA Audit of Corrective Action Program Audit Number: 85-CA-2 Date Audited: September 17, 1985 - October 25, 1985 Activity: QA Audit of Corrective Action Program Audit Number: 85-QC-2 Date Audited: June 10-20, 1985 Activity: QA Audit of Quality Control Program Audit Number: 85-QC-3 Date Audited: September 3-16, 1985 Activity: QA Audit of Quality Control Program Audit Number: 85-QCA-1 Date Audited: February 18-28, 1985 Activity: The QC Section and other QC programs being implemented by other activities. An overall assessment of plant activities based on the corporate QA program, Project QC Section, and other related QC activities being implemented by activities other than the Quality Control Sectio Audit Number: 85-MA-1 Date Audited: February 11-25, 1985 Activity: Quality Assurance Audit of Maintenance Audit Number: 85-MC-2 Date Audited: June 11-28, 1985 Activity: Quality Assurance Audit of Material Controls
_ _ _ _ _ . _ . _ _ _
 
        . _ ._. ..
  .
 
'
 
Audit Number: 86-CA-1 Date Audited: March 3,1986, April 1-4,1986, April 7-8,1986, April 11, 1986, April 21, 1986 l
Pursuant to the review of the above audit reports and the corrective actions initiated for identified deficiencies, the inspectors concluded that the audit program and its implementation appears to be adequat The inspectors determined that licensee management had established a QA trend program. The objective of this program is to provide appropriate management with advanced warnings of real or potential problems as indicated by possible adverse trends. Problems identified through this process are then further evaluated by assessments, audits, surveillance or investigatio The data base of the QA Trend Program consists of the results of all QA audits, QA surveillance findings (except housekeeping), NRC findings, and INP0 evaluations. The data base sample covers three months time interval All data collected is classified in accordance with the following code:
0A TREND CODES Principal Trend Code  Description E    Personnel Error C    Construction / Installation D    Design j    Q    Programmatic The principal trend codes are to be applied based on the following definitions:
E - Personnel Error - An error made when a written procedure was not followed or because personnel did not perform in accordance
 
with accepted or approved practic C - Construction / Installation - This classification is assigned to problems reasonably attributed to the construction or installation
. of a system, component, or structure. For example, failure to comply with approved drawings and/or designs which result in an
    "as-built" condition which deviates from the intended conditio This may be shown by physical inspections showing deviations or by the failure of the component to meet its functional test require-
,    ments.
 
!
i
._- . - . _ _ .______ . _ _ _. . _ _ . . _ _ _ _  _ _ _ . _ _ _ _ _ _ _ . . _ _ _ _ . _ _ _ . . _ . _ _ ,.. _ ____ __ _ ,, _ .
 
0 - Design - This classification is assigned when the problem is reasonably attributed to design of a system, component, or structure. For example, problems that are traced to such things as incompatibility of materials, loss of design control, or choice of components which fail to meet specified functional requirements or performance specification Programmatic - This classification is assigned to problems caused by the failure of management or management systems (e.g.,
breakdown in administrative controls, preventative maintenance program, surveillance program, or quality assurance controls).
 
This would include FSAR commitments not met or implemented and license conditions not implemented.
 
Each principal trend category is further sut' divided into cause code sub-categories.
 
A complete graph of each principal trend code is then prepared using the guidance delineated in procedure QA-05-20. Additionally, reviews and verification of potential problems as depicted by the graphs are performed in accordance with the administrative controls described in this procedure.
 
The following documents were reviewed by the inspectors in connection with review and assessment of the QA trend progra Interoffice correspondence from P. E. Fornel-QA Site Manager to D. S. Read - General Manager RE: NRC/QA Item Trend Analysis, First quarter 1986, dated March 4,198 Interoffice correspondence f rom L. C. Byrnes to D. S. Read, '
Subject: Supplement to the First Quarter 1986 Trends Report, Interoffice Correspondence from Q. M. Fraser - Acting Site QA Manager to D. S. Read-General Manager Quality Assurance, RE:
NRC/QA Item Trend Analysis Second Quarter 1986, dated June 6, 1986.
 
The Second Quarter Trend Report identified an increased number of personnel errors caused by a lack of attention to detail or failure to follow procedure. These personnel error findings were identified in the areas of design change requests, emergency plan, plant operations, fire protection, measurement and test equipment and corrective actions. In discussions with licensee management, the inspectors were informed that the increase in personnel error can be attributed to fatigue and preoccupation with Unit 1 outage activities, and the time constraints associated with completing the outag . - - - . . . - _ _ . -
  .
*
:
 
l,
!
i The inspectors were further told that the majority of personnel errors l  were made by contractor personnel. In response to this finding, licensee management has instituted administrative controls which require vendor activities onsite to be in accordance with all Plant Hatch procedures and the QC program. A selected list of procedures requiring the attention of vendor / contractor personnel prior to any
,
onsite activities have been prepared and is included as an Appendix to
'
all requests for procurement of service The inspectors concluded that the licensee's QA Trend Program meets the requirements of licensee canmitments delineated in ANSI 18.7-1976, and appears to be effectively implemented. Personnel errors identified may be indicative of a generic problem of loss of control of personnel / procedural interface requirements within the " work control" process. This assessment is substantiated by the fact that numerous new procedures are presently being prepared by the Procedures Upgrade Program (PUP). Additionally, the requirement to ensure that plant f  personnel have been trained to the new procedures have not yet been fully accomplished. The inspectors commented favorably however, upon
:  licensee management arrangements to have plant personnel participate in
;  the validation of the new procedure NRC Inspection Findings l  The inspectors reviewed licensee program documents and conducted interviews with licensee management to assess the status of NRC open
,
,
being drafted to establish the criteria for the audit. The licensee committed to implement a procedure for the Operating Experience Program review by February 28, 1987. The licensee argues, however, that although the event occurred, there were no actual or potential safety consequence NRC Evaluation The NRC does not agree with the licensee's statement that the violation had no  l actual or potential safety consequences. The NRC staff noted that in April of  ,
items. The following document was reviewed in connection with this effort:
1 1985, the licensee was cited for a failure to provide operational experience  I feedback training to maintenance personnel (321, 366/85-07-03). The licensee's  i responses indicated that maintenance personnel would be added to the procedure  l governing operational feedback training (HNP-911), and that operating  l t
Interoffice correspondence LR-QAM-004-0786 from 0. M. Fraser,
experience feedback training would be provided to maintenance personne l During the current inspection, over a year later, it was determined that this  )
*
required training had still not been implemented for maintenance personne In '
Acting QA Site Manager, RE: Summary of Plant Hatch NRC/QA Status
addition, the failure to perform the audits of the functioning of the program at all levels, as cited above, resulted in this deficiency continuing for an
;    dated July 2, 198 The inspectors determined that for the period June 1 through June 30, i  1986, two NRC inspection reports were received by the QA department
,   that resulted in eight items being opened and two items being closed.
 
l   Also, of the present 100 NRC open items, 67 are awaiting NRC review
;
after completion of corrective action and 33 have made satisfactory
;  progress for the mont The inspectors concluded that adequate corrective actions are being
,
,
extended period of time, and in less than timely corrective action to the 1985 1 violatio ;
implemented by licensee management regarding NRC inspection finding Internally Identified Problems The inspectors reviewed the following documents to assess the status of
.
licensee internally identified problems documented as Deficiency l  Reports (DR):
.
Interoffice Correspondence No. LR-ENG-006-0786, from T. R. Powers  t RE: DR Report No. 86-2
  - - _ .. _ . _ __ _ _ _ . __ _ _ _ ,. _ _ _ _ _ _ _ _ _ _ _ _  _
_ _ _ . _ _ _ _ _ _ _


l
.
.
 
Interoffice Correspondence No. LR-MGR-058-1085, from H. C. Nix RE:
Deficiency Report Trend Analysis Report (No. 85-6)
Interoffice Correspondence No. LR-MGR-102-0885, from H. C. Nix RE:
Deficiency Report Trend Analysis Report (No. 85-5)
Deficiency Report No. 86-2 indicated that through April 30, 1986, the number of DRs was 93 This showed an increase of 352 DRs over the number of DR's generated for a similar period in 1985. Licensee management attributes this increased rate of DRs to the Unit 1 outage (November through May), when more DRs were generated because of work activities associated with plant shutdow The breakdown of the DR's identified in DR Report No. 86-2 are as follows:
219 were due to component failure 214 were due to design, manufacturing, construction 249 were due to personnel error 118 were due to defective procedure 5 were due to external causes 94 were due to other cause Of these, 68 were found not to be valid deficiencies Note that 35 DRs had inadequate responses on which to base cause code assignments. These responses should be upgraded before the next report is issued.
 
I
! Figure 13A, Personnel Errors by Responsible Department, identified the
! three largest contributors as 29.9 percent unknown; 25.2 percent due to maintenance; and 18.9 percent due to contractors. The large percentage (29.9 percent) identified as unknown may be indicative of inadequacies in licensee's root cause analysis of personnel error Figure 14A, Personnel Errors by Cause Code, identified the three largest contributors as 42.6 percent due to poor craftmanship; 2 percent as failure to follow procedures; and 14.3 percent as failure l to use reference materia The inspectors concluded that the information provided by the DR trend
'
'
        !
analysis is similar to that obtained from the OA trend program. The DR trend analysis further corroborates the inspector's previous assessment of a generic problem involving a breakdown of the controls associated with the personnel / procedural interface requirements within the work l
        :
l
.- - -. -- . - - - - - - - -- - - -.. .  - - -- -


_  _ _ _
_
  .
  .
  .
  .
control proces Corrective actions that are teing implemented by licensee management include additional training of plant and contractor personnel, and a Procedures Upgrade Program which is presently ongoing. These two issues are discussed further in other parts of this repor g. Operational Events The inspectors reviewed the following documents to assess the status of licensee operating events that are identified as licensee event reports
  '
  '
Enclosure  4 NRC Conclusion The NRC considers that the licensee's statement that the violation had no actual or potential safety consequences was not appropriate. The implementa-tion of program features which assure quality and the prompt correction of deficiencies are considered essential to plant safet Restatement of Violation C 10 CFR 50, Appendix B, Criterion V, and the licensee's accepted QA program (HNP-2, FSAR-17, Section 17.2.5) require that activities affecting quality shall be prescribed by procedures of a type appropriate to the circumstances and shall be accomplished in accordance with these procedure Hatch administrative control procedure 50AC-MNT-001-0S, Maintenance Program, Step 8.6.2, requires that: the results of all inspections and tests performed shall be documented and further action based on results and the documentation will become part of the MW0 packag Contrary to -the above, the licensee failed to document the results of the preliminary leak rate test associated with MW0s 1-86-5083 and 1-86-5084 concerning repairs on drywell vent valves 1T48-F319 and IT48-F32 Summary of Licensee's Response The licensee stated that the cause of the violation was attributed to i unclear procedural wording that exceeds the intent of the procedure. The licensee also argues the following point The procedure was not intended to transform diagnostic tests into formal tests for which documenta-tion was required. It was a standard practice for the local leak rate test (LLRT) personnel to perform a preliminary diagnostic test after valve repair and reinstallation prior to completing reassembly (i.e., installation and hookup of valve operator). This diagnostic check was intended to minimize both labor and radiation exposure by early identification of problems, prior to
(LER):
Interoffice Correspondence No. LR-ENG-026-0286 from T. R. Powers, RE: LER Trend Report No. 86-1 Interoffice Correspondence No. LR-MGR-004-0186, from C. T. Jones, RE: LER Trend Report No. 85-4 Interoffice Correspondence No. LR-MGR-024-1258, from C. T. Jones, RE: LER Trend Report No. 85-3 Interoffice Correspondence No. LR-MGR-095-1185, from D. T. Jones, RE: LER Trend Report No. 85-2 Licensee LER Trend Reports were developed by evaluating each LER and i assigning it to one of the following cause codes:
l 1  A Personnel error
"
B Design, manufacturing, construction C External cause D Defective procedure E Component failure X Other Each cause code was then further broken down into areas of signifi-cance, to identify, as far as possible, the reasons for component failure, personnel error, etc., root cause analysis may not be
,
,
formal LLRT. In the cited case, the diagnostic test was performed subsequent
performed to ascertain any single root cause for all reported problems.
! to the repair of valves 1T48-F319 and 1T48-F320. Excessive leakage was apparent, and the valves were removed, repaired and reinstalled as part of the process of correcting the observed leakage. The fact that a diagnostic test was performed was not noted on the MWO continuation; thus, the inspector, as noted in the inspection report, could not determine why the valves had been remove The licensee stated that on November 3,1986, a memo was issued requiring that l diagnostic tests be noted on an MWO continuation sheet. The licensee concluded ,
 
; that the Maintenance Program procedure (50AC-MNT-001-0S) was over inclusive and l stated that it would be revised to differentiate between the documentation of required tests and those which are being performed for diagnostic or other purposes. The licensee committed to complete this revision by January 10, 1987.
< Depending on the nature of a problem and the availability of manpower
. and equipmen LER Report No. 86-1 trended events which occurred from January 1984 through December 1985. Significant parts and trends identified in this
; report are as follows. There was an increase in the rate of LER
' occurrence in that the monthly rate of LER's for 1985 was 71/4 as compared to Si for 1984. Additionally, LER's attributed to personnel errors showed a constant trend of 32% and 33% over the time spans from November 1, 1984 through October 31, 1985, and January 1, 1985 through December 21, 1985, respectivel _ _ _ _ _ . _ - _ _ - . _ - _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ . ___
 
.


, With regard to Violation "C", the licensee argued that the violation had minor,
The inspectors determined that the data provided by the above trend report is consistent with the evaluation independently arrived at by the NRC by analysis of reported LERs. The inspectors also concluded that the data from the LER Trend Report is consistent with that provided by the QA Trend Program and the Deficiency Trend Analysis Report This is based on the observation that a generic problem involving lack of control of personnel / procedural interface require-ments within the " Work Control Process" appears to exis Component failure trends showed that 70 percent of the 21 Unit 1 LERs resulting from component failures were initiated by probicms within 5 systems. For Unit 2, 65 percent of the 19 LERs resulting from compo-nent failures were initiated by problems within three systems. Figure 16A, Lurrent Component Failure Causes, shows 32 percent of the causes as unknown. This may be indicative of inadequacies in the licensee's root cause analysis precess. Additionally, the large percentage of component failures initiated by unknown causes may also be interpreted as indicators of problems within the work control process as it per-tains to plant / personnel interface requirements. Corrective actions that are being implemented by licensee manageinent such as upgrade of personnel training, and upgrade of procedures are discussed in detail elsewhere in this repor Within this area two violations were identified and are discussed in the following paragraph Failure to Follow Procedure A sample size of 30 closed out safety related maintenance work orders (MWO) were reviewed by the inspectors to verify procedural compliance I with the QA program. Three MWO's were identified which indicated that maintenance activities were conducted in a manner such that deficien-cies found during performance of these work activities were not promptly documented, evaluated, and corrected. Maintenance Work Order No. 1-86-3998 required performing the overspeed test on the HPCI turbine. The test acceptance criteria was not achieved, and contrary to procedural requirements, maintenance was performed to correct the deficiency. Documentation and evaluation of the deficiency for root cause analysis was never performed for this identified proble Maintenance work order No. 1-84-4872 required that a turbidimeter be inspected and repaired to facilitate calibration of the instrument. It was discovered that the problem was not a defective instrument, but rather incorrect calibrating guidance contained within the calibration procedur Contrary to the QA program requirements, a deficiency report was not prepared to document and initiate corrective actions for this identified deficienc Similarly, MWO No. 1-86-3426 required that the fuel pool cooling holding pump be investigated and repaired to correct a problem with the pump failing to start when the control switch was placed in auto or manual position. Investigation revealed that the thermal overload relays had tripped and the motor full-load running current was not at the rated value. A deficiency report to
; if any, safety significanc The licensee requested that, consistent with the Commission's Enforcement Policy (10 CFR Part 2, App. 2), that the cited Violation be recategorized as a Severity Level V. The licensee stated that the
:
I
- _ - - ,
  . ... -, . - - - . - - -
      - ,


  .
  .
document and initiate root cause analysis of this problem was not prepared, and no further actions were taken. These failures to adhere to the requirements delineated in procedures 50-AC-MNT-001-05 and 10 AC-MGR-004-0 to assure that deficiencies identified during mainte-nance activities are documented, evaluated, and promptly corrected is identified as violation (50-321,366/86-22-06).
h. Operational Experience Feedback The Maintenance Manager indicated at the entrance meeting that operational experience feedback information does not go past the maintenance foremen (i.e., to individual mechanics), but is instead inserted into procedures as required. This method of dispensing operational experience may be adequate in some circumstances, but this approach does not fully meet the intent of NUREG 0737 Item 1.C.S. In some cases, it is essential that actual formal training be provided to the individuals responsible to emphasize the reasons for the changes and the safety significance of errors. This training can also be in the fonn of a discussion of the sequence of events and contributing causes, or how a particular event at another plant relates to mainte-nance at Hatch. Required reading can also be utilized in certain applications as a method of dispensing operational feedback to maintenance personnel as it is in operations. Operating experience needs also, in some cases, to be incorporated into formal training programs such as those presently being implemented at Hatch in the maintenance area. Placing significant operating experience into the training and retraining programs will ensure that new personnel can also avoid repetition of errors and events that have occurred within Hatch or at other facilities. The key to ensuring that significant operational experience is distributed adequately to all affected personnel is a formal documented program, with an adequate feedback mechanism to ensure all affected individuals have received the training or reviewed the material. The licensee received a previous violation (321,366/85-07-03) for failure to establish an operational experience feedback program for maintenance personnel to ensure that they are informed and training programs upgrade Due to an apparent continued reluctance by the licensee to establish a formal program, in addition to the informal newsletters and bulletins, this violation could not be closed by the inspector NUREG-0737, Item I.C.5 delineates the requirements for preparing and implementing procedures governing feedback of operating experience to plant staff on or before January 1,1981. Paragraph I.C.S.7 further states that periodic internal audits shall be performed to assure that the feedback program functions effectively at all levels. Engineering procedure 30AC-0PS-005, implements this requirement and requires that audits of the functioning of the operational feedback program at all levels be audited on a biannual basis. In discussions with licensee management the inspectors determined that an audit of the Operating Experience Feedback Program was conducted by the QA organization and i the findings are documented in Audit Report No. 85-TR- This audit
{
__ __
'
.
.
Enclosure  5 literal words of adrinistrative control procedure 50AC-MNT-001-0S were over inclusive, requirirg diagnostic steps if characterized as a " test" to be documente The licensee committed to correct the administrative control procedure so that only tests which are used to determine operability or to meet operability criteria are documente The licensee argued, however, that diagnostic tests, such as the one identified by the inspection report, are akin to visual observations of conditions and are not intended to be formally documente NRC Evalut. tion The NRC staff does not agree with the licensee's statement that formal documentation would be limited to only tests that determine operability or meet operability criteri The American National Standard (ANS) -
3.2/N18.7-1976, which the licensee is committed to in their Quality Assurance Program (HNP-2, FSAR-17, Appendix A.33), Section 5.2.7, Maintenance and Modification, requires that means for assuring quality of maintenance and modification activities (for example, inspections, measurements, tests, welding, heat treatment, cleaning, nondestructive examination and worker qualification in accordance with applicable codes and standards) and measures to document the performance thereof shall be established. ANS-3.2/N18.7-1976, defines inspections as examination, observation or measurement to determine the conformance of materials, supplies, components, parts, appurtenances, systems, personnel performance, procedures, processes or structures to predetermined requirements. Additionally, the standard defines testing as performance of those steps necessary to determine that systems or components function in accordance with predetermined specification As per the committed standard, tests, inspections, observations and/or as the licensee states diagnostic tests are means for assuring quality'of maintenance and require documentatio NRC Conclusion The NRC accepts the licensee's statement that this single event has minor safety significance, therefore, the severity level will be changed to Restatement of Violation D 10 CFR 50, Appendix B, Criterion XII, and the licensee's accepted QA program (HNP-2, FSAR-17, Section 17.2.12) require that measures shall be established to assure that tools, gauges, instruments, and other measuring and testing devices used in activities affecting quality are properly controlled, calibrated, and adjusted at specified periods to maintain accuracy within necessary limit Contrary to the above, the licensee failed to assure that torque multiplier number 391A268517 was properly calibrated or adjusted to maintain accuracy within necessary limits when used in conjunction with MW0s 1-86-5083 and 1-86-508 l l
l
      ;


..
was not intended to meets the requirements for audit of the program delineated in procedure number 30 1AC-0PS-003. This audit was a small part of an overall QA audit, and addressed only the fact that there was a procedure in place for controlling operating experience feedback to Operations personne The inspectors were informed that audits of the feedback of operating experience program performed under the administrative  controls This failureofto  the
.
!  Engineering Department have never been conducte conduct periodic audits as required by NUREG-0737 paragraph I.C.S.7 and Procedure 30 AC-0PS-003 is identified as violacion (50-321, 366/86-22-07).
l Enclosure    6 Summary of Licensee's Response The licensee denied the violation. The licensee argued that calibration of torque multipliers is an issue which had been previously identified by the licensee as a result of ongoing evaluations related to valve maintenance and was being resolved at the time of the inspection. The licensee stated that corrective steps included the review of torquing practices at other plants and issuance, on November 6,1986, of a Standing Order requiring the use of direct reading torque indicators with torque multipliers or, when a direct reading indicator cannot be used, a calibrated torque multiplier or other engineering means for confinning proper torqu In addition, a general maintenance procedure on torquing (51GM-MNT-033-05) had been prepared and was in the review, approval, and validation process. The licensee stated that this procedure contained requirements similar to those in the above Standing Order. The procedure was scheduled to be issued by December 10, 198 NRC Evaluation At the time of the inspection, the week of July 28, 1986, the licensee did not indicate or provide any ongoing evaluation related to the use and/or calibra-tion of torque multiplier If the evaluation was ongoing, the continued use of non-calibrated test equipment (torque multipliers) until November 6, 1986, a three-month period, appears to be marginal for timely corrective action However, the NRC accepts the licensee's statement that the item was licensee identifie NRC Conclusion For the above reasons, the NRC staff concludes that the violation will be withdrawn. Our records will be adjusted to reflect the withdrawa ,
 
t
1. Control of Contract Maintenance Personnel Due to previous allegations and problems involving contract personnel and valve maintenance, the licensee offered to make a concerted effort
  , , . _ . . - . - . _ . - .- , -. - - _ - ~
.
'
to upgrade Georgia Power controls over contractors on site. The
'  licensee indicated that contract personnel would now be required to provide documentation of training and qualifications related to work tasks to be performed. Previously, the contracting firms certification l  of individual qualifications had been considered adequate. The contract personnel are also being required to complete at least one week of site specific training. In addition to GET training areas such as radiation protection and security, this training emphasizes  This the time need is to follow procedures in performance of maintenance task also utilized to familiarize each contract maintenance individual with the procedures relating to the specific tasks that he will be preformin The licensee indicates that the number of Georgia Power personnel directly supervising the work of contract personnel will be substantially increased. The total number of contract firms  A supplying review of maintenance personnel will be reduced for added contro the personnel error rates by the inspector, involving both contract and Georgia Power personnel, indicated that these additional controls were definitely warranted. The impending refueling outage should provide an indication whether these newly established contractor controls will
;
be effective.
 
!
;
!
!
_ -
  - - _ - . _ . - --_ ,. _ _ . . _ _ _ . _ _ _ . _ _ _ _ _ _ _ . _ . _ . _ , _ _ _ _ . _ _
}}
}}

Latest revision as of 00:24, 19 April 2021

Insp Repts 50-321/86-22 & 50-366/86-22 on 860728-0801. Violations Noted:Failure to Perform Matl Verification & to Adequately Document & Verify Matl Used During Valve Replacement
ML20215K584
Person / Time
Site: Hatch  Southern Nuclear icon.png
Issue date: 09/26/1986
From: Stadler S, Wilson B
NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION II)
To:
Shared Package
ML20215K512 List:
References
50-321-86-22, 50-366-86-22, NUDOCS 8610280220
Download: ML20215K584 (32)
v • d • e


Text

p ucg UNITED ST ATES

/ o NUCLEAR REGULATORY COMMISSION

[ n REGION li g ,, j 101 MARIETTA STRCiET. *I '* ATLANTA, GEORCI A 30323

%...+/ +

Report Nos.: 50-321/86-22 and 50-366/86-22 Licensee: Georgia Power Company P. O. Box 4545 Atlanta, GA 30302 Docket Nos.: 50-321 and 50-366 License Nos.: DPR-57 and NPF-5 Facility Name: Hatch 1 and 2 Inspection Conducted: July 28 - August 1, 1986

'

Inspectors: M . .

i[M[/4 S.~ Stadler ~ Date Signed H. O. Christensen W. K. Poertner G. A. Schnebli C. F. Smit J.,B. Brady J. D. Smith Accompanying Personnel: ' V. L. Brownlee, Branch Chief Approved by: hh, w w #/)68$

B. Wilson, Acting Chief, Operational Programs Date Signed Division of Reactor Safety SUMMARY

, Scope: This routine, special announced inspection was conducted in the areas of maintenance programs, implementation, and corrective action Results: Four violations were identified.

8610280220 861017

,

PDR ADOCK 05000321 0 PDR I

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

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

REPORT DETAILS Persons Contacted Licensee Employees

  • Harvey Nix, General Manager
  • Tom Seitz, Manager Maintenance
  • Lewis Summer, Manager of Operations
  • Tom Greene, Deputy General Manager
  • Zachary Wahab, Engineering Supervisor
  • Roger W. Zavadoski, Manager, Health Physics / Chemistry
  • Jimmy Wilkes, Manager, Special Projects
  • D. F. Moore, Nuclear Training Coordinator
  • 0. M. Fraser, Acting Quality Assurance Site Manager
  • B. K. McLeod, Manager Maintenance and Outage
  • P. R. Bemis, Manager Engineering General Offices
  • B. R. Phillips, Training Supervisor
  • O. Porter, Assistant Project Engineer
  • R. A. Glasby, Project Manager
  • A. Vest, PWPS Manager, Onsite
  • P. A. Robertt., Acting Independent Site Engineering Group Supervisor
  • S. H. Chesnut, Nuclear General Engineer
  • R. K. Moxley, Quality Assurance
  • S. J. Bethay, Acting Regulatory Compliance Superintendent
  • C. R. Goodman, Acting Regulatory Compliance Supervisor
  • G. A. Goode, Acting Manager Engineering i *C. T. Moore, Manager Training Other licensee employees contacted included engineers, technicians, operators, mechanics, and office personne NRC Resident Inspector
  • P. Holmes-Ray (SRI)
  • Attended exit interview Exit Interview

The inspection scope and findings were summarized on August 1,1986, with those persons indicated in the paragraph above. The inspector described the areas inspected and discussed in detail the inspection finding No dissenting comments were received from the license The licensee did not identify as proprietary any of the materials provided to or reviewed by the inspectors during this inspectio _)

_

3. Licensee Action on Previous Enforcement Matters This subject was not addressed in the inspectio . Unresolved Items Unresolved items are matters about which more information is required to determine whether they are acceptable or may involve violations or deviation Unresolved items identified during this inspection are discussed in paragraph . Maintenance Training Program The mechanical, electrical, and instrumentation and control (I&C) training courses are the last three courses to be submitted for INP0 accreditation by the~ licensee. The other six courses, licensed operator, non-licensed operator, requalification, shift technical advisor (STA), radiological protection, and manager and technical staff, were already INP0 accredite The inspector reviewed the Self Evaluation Reports (SERs) associated with the three remaining training programs to be accreoited, and the status of development. This review indicated that the licensee has prepared in-depth for the accreditation of these courses. A full job task analysis (JTA) was performed for each of the three maintenance positions, and all of the supporting materials such as texts and lesson plans will be completed prior to submittal for accreditatio In addition, the licensee has solicited heavy participation (up to 90 percent) by personnel from each of the three maintenance groups in the development of the JTA, courses, and related training material Involvement by job incumbents in the development of performance based training not only increases the technical accuracy and relativity, but can substantially increase the course credibility and i acceptance of the training by the various group At the time of this inspection, the status of development of these last three courses to be accredited were as follows:

-

Instrument and Control Technician: 119 of 141 modules completed or 84.4 percent

-

Electrician: 58 of 97 modules completed or 59.8 percent

'

- Mechanic: 45 of 93 modules completed or 48.4 percent The licensee indicated that development was on an accelerated schedule and

,

that all three courses were scheduled to be completed and submitted to INP0 by September 30, 1986 In an effort to upgrade the proficiency and performance of the Maintenance Group, the licensee is not only requiring that new hires and transfer personnel attend these courses, but also that all job incumbents complete the trainin Completion of the full training course by incumbents is projected to take up to three years for some individuals, and early indications are that there may be a substantial attrition rate due to the

__ . _ . increased level of training expectatio To counteract this expected attrition rate, the licensee is offering financial incentives for successful completion of the trainin Evening courses in remedial mathe-matics are also being offered on a voluntary attendance basis for all maintenance personnel. As pointed out in their June 12, 1986, presentation to the Region, the licensee has also increased the acceptance level criteria for new maintenance personnel. The inspector reviewed this upgrade in acceptance standards and noted that several significant changes had been made. A previously utilized written aptitude test was replaced in June, 1986, with the more comprehensive and widely used MASS test. The MASS test is administered to all non-transfer applicants for the mechanic, electrician, or instrument and control technician position The licensee believes the new test to be more relevant to the positions and a better indicator of potential success, as well as being more difficul In addition, the licensee discontinued the differential grading of written aptitude examinations for minority versus majority candidates in December, 1985. All applicants for maintenance positions are now subject- to the same acceptance criteria. The licensee has also added a skills test for mechanic and electrician applicant This test was developed to assess the

" hands-on" skill levels and was validated utilizing experienced job incumbents. Each skills test takes about three hours and requires the applicant to demonstrate three basic skills applicable to the mechanic and electrician position The instrument and control technician, which is a non-union position, requires no skills test for applicants due to added experience requirements. The licensee estimated that this skills test would screen an additional 50 percent of the applicants for the mechanic and electrician position To facilitate the training of all maintenance personnel, including new hires and incumbents, the licensee is attempting to establish a shift schedule which will allow regular rotation through trainin A twelve-hour shift schedule was apparently rejected by the union, and there was inadequate staffing to support a six-shift schedule. Resolution of this schedule problem will be a key to ensuring that maintenance training is implemented on a timely and continuing basis. To allow maintenance training and qualification to occur on all shifts the licensee has appointed a full time training coordinator and assigned training evaluators on each shift. The evaluators will conduct on-shift evaluations and qualification checkoffs on specific job tasks for maintenance personnel in training. Although there are long term plans for a computer qualification card system, individual task qualifications at the time of this inspection were being tracked on new qualification card With increased maintenance staffing and the training of all incumbents, maintenance personnel will be in many different stages of training and qualified only on certain job tasks for some time. Considera-tion should be given to expediting the implementation of the computer qualification syste With adequate access terminals, this would allow maintenance supervisors to quickly access whether an individual is qualified for a particular task, or all individuals qualified for the tas This system should also help ensure that unqualified individuals are not assigned to perform maintenance tasks.

!

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

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

The inspector toured the hands-on training laboratories established in the

'

training building for the Maintenance, Electrical, and I&C groups. These facilities had been inspected in the past and determined to be exemplar In preparation for INP0 accreditation and the implementation of the new performance based training programs, these superior facilities were being upgraded even further. The electrical laboratory contained a number of plant specific work / training stations and mockups. An electrical breaker is set up with all terminal connectors contained in a separate box. Trainees can disassemble and work on the breaker or trouble shoot problems in the terminal box. Breaker logics are also mimicked and instructors can insert logic problems utilizing a set of toggle switche Once the trainee has completed trouble shooting and repairs, the breaker can be operated to determine success or failure. Other training stations reviewed by the inspectors included duplication of various electrical relays, and a large valve with a Limitorque operator for trouble shooting and rebuilding. The electrical training course also includes 40 hours4.62963e-4 days <br />0.0111 hours <br />6.613757e-5 weeks <br />1.522e-5 months <br /> of training in the inter-pretation and use of electrical drawings, procedures, vendors manuals and other documents and the related symbols, acronyms, and terminology. The I&C Laboratory also was equipped with a large number of installed work stations and mock-ups. Where generic training stations or panels were purchased from vendors, they were modified to be plant specifi All training staff personnel interviewed indicated that plant management had been extremely cooperative in providing various training aids, such as pumps, valves, and breakers, so that the labs could be well equipped and plant specific. The

! I&C lab also had the capability to establish complete control loops for pressure, level, or flow instrumentation for trouble shooting and trainin This laboratory was particularly impressive and should allow duplicating most any major I&C repair, trouble shooting, or surveillance testing require-ments. The I&C training staff has also recently begun utilizing the plant simulator to train I&C technicians on responses of control room instrumen-tation such as neutron and power level monitoring, the rod worth minimizer l rod sequence control system, rod block monitor, and reactor water level controls. This is a relatively unique approach and could prove to be very l beneficial training. The maintenance laboratory was also very well equipped with various types of pumps and valves available for maintenance and trouble

shooting. One pump was connected to a water storage tank so that maintenance trainees could conduct pump curve analysis following repairs and maintenance to the pump. This capability provides positive visual feedback to the

! trainees on their performanc We mechanical lab also emphasizes the I administrative aspect of complet r.) maintenance assignments. Trainees are

'

required to fill out and proce.s work requests, to fill out material and tool requisitions, and to complete and document post maintenance functional testing results.

! In summary, the corporate and plant support of maintenance training appeared I

to be comprehensive and logically developed, and the training facilities l probably among the best available anywher If the programs are implemented l as designed, and the union and scheduling problems resolved, the level of proficiency among maintenance personnel, and ultimately maintenance performance, should definitely increase. Since these training courses had been implemented only two weeks prior to this inspection, the implementation and effectiveness will be assessed during subsequent inspections.

l

,

j

6. Procedures Upgrade Program The procedures upgrade program (PUP) has been in existence at Hatch for the past three years. The phase I upgrade which concluded late in 1985 was apparently much less than effective, and did not result in any significant upgrade of procedures. Late in 1985, the primary contractor associated with Phase I was terminated and a new procedures upgrade group was formed. This new PUP group consists of approximately 60 full time personnel devoted to the upgrade of Hatch procedures, and this revised effort is being desig-nated as Phase II. To help ensure the success of Phase II, the licensee has centralized the PUP effort and is providing for increased Georgia Power involvement in the process. The licensee's objectives in Phase II of PUP are basically to ensure that all Hatch procedures are revised to be technically accurate, user friendly, and clear in intent to every qualified procedure user. To ensure that these objectives are met in Phase II, the licensee has implemented what appears to be a very rigid quality control program for procedure revisions. Plant management has provided nine full time personnel to assist in the procedure development, four from Maintenance and five from Operations. These personnel should not only assist in ensuring the technical accuracy and usability of the procedures, but their involvement in the development should instill a higher level of user confidence in the final produc In addition to the use of these subject matter experts, the licensee has incorporated several other controls in the development stage to ensure technical accuracy. The minimum qualifications for procedure writers has been procedurally defined as requiring 2 to 3 years in the nuclear field, a good technical understanding of boiling water reactors (BWRs) and procedural discipline, and previous procedure writing experience. For procedures related to systems and equipment, the procedure writers are required to cross check the vendor's manuals to ensure that applicable recommendations and requirements are incorporate The systems and equip-ment are also " walked down" as required to ensure the technical accuracy of terminology and locations versus actual plant conditions. The draft procedures are also verified against approved plant drawings to ensure accuracy. Finally the commitment tracking system is checked to ensure that all required commitments are incorporate The draft procedure review process is also tightly controlled and i procedurally defined. The initial reviews are performed internally by the technical advisors and by at least one other procedure writer. The draft procedures also undergo an as low as reasonably achievable (ALARA) review by Health Physics, and review and approval by Regulatory Compliance, QA and QC, Procedure Review Group (PRG), and the Plant Review Board (PRB). A new validation process which was incorporated into the procedure revision approval process approximately tWo months ago should add significantly to the technical accuracy, as well as the procedure credibility among users.

,

This validation process requires members of the user groups, other than those involved in the development process such as Maintenance or Operations

!

l l

personnel, to validate the new procedure by actual or simulated use. The methods of validation utilized, listed in the order of preference, include the following:

- actual performance of the procedure

-

use on the plant simulator (Operations procedures)

-

use on plant specific training mockups such as pumps and control stations

- walkthrough of the procedure without actually performing any work or evaluations

- in office comparison to the original procedure In response to inspectors concerns, the licensee indicated that the actual performance would always be the primary method of procedure validation subject to system or equipment availability and plant conditions. In addition, the in-office reviews against the original procedures would be limited to administrative type procedures, not operating, maintenance, or surveillance procedures.

The licensee intends to revise a total of approximately 5,000 procedures and had accelerated the schedule about one month before this inspectio All mechanical and electrical procedures (Maintenance Department) are scheduled to be completed within the next 6 months. The Instrument and Control (ISC)

procedure revision is not scheduled to be started until after the mechanical and electrical procedures are completed. The licensee based this schedule on the assumption that the I&C procedures were in better condition, requiring less priority. The inspectors noted that there were a number of I&C personnel errors in the last two years, and the events should be analyzed to ensure inadequate I&C procedures were not at faul t. If procedures were a large contributor to the error rate, this priority should be reevaluated. The general overall status of the PUP effort at the time of this inspection wasas follows:

-

840 drafts completed in 1986 of which 376 were in the maintenance area

- 46 percent of mechanical and electrical procedures completed

- 21 percent of all procedures completed

- 15 percent of I&C procedures completed

- only 4 procedures had been validated In sunnary, Phase II of PUP appeared at this stage of completion, to be an effective, ambitious, and well controlled effor Support by plant and corporate management was very good, including manpower for development and review. This manpower connitment needs to be extended to the validation process to ensure the continued quality of the procedures (only 4 have been validated to date). The confidence level in, and acceptance among procedure users of the new procedures appeared to be very high. Some confusion seemed to exist,'however, in how to determine a new procedure from the old in that the numbering system is identical. This confusion could cause employees to attribute inadequacies in an old procedure to a new procedure, and reduce

. -- - - .-_

7 confidence in the upgrade program. If this problem can be addressed, and the present level of development and review not downgraded to expedite development, the Hatch Procedures Upgrade Program could prove to be one of the more successful efforts of this typ . Maintenance Program (62702)

The inspectors reviewed the following procedures:

50AC-MNT-001-05: Maintenance Program i 50AC-MNT-007-0S: Preventive Maintenance Program

! DI-REG-10-0286N: Nuclear Plant Reliability Data System 51GM-MNT-007-0S: Control of Lubricants 6 30AC-0PS-005-0S: Temporary Bypass, Jumper, and Lifted Lead Control DI-MNT-02-1085: Maintenance History and Trending Program DI-REG-08-1285N: DR and LER Trending Program

51GM-MNT-002-0S: Maintenance Housekeeping and Tool Control i

j 40AC-ENG03-0: Design Control 40AC-Eng-009-0S: Control of Special Processes

.

30AC-0PS-002-0S: Plant Housekeeping and Cleanliness Control i

'

30AC-0PS-001-0S: Equipment Clearance and Tagging 50AC-MNT-002-0S: Control of Measuring and Test Equipment 45QC-QCX-002: Quality Control Inspection Plans 53PM-MON-001-0S: Vibration Analysis

'

53PM-MON-002-0S: Lube oil Analysis I

42EN-INS-001-0S: ISI Pump and Valve Operability Program 40AC-FPX-001-05: Fire Protection Program The inspectors conducted a review of the above procedures to ascertain that the licensee has implemented a maintenance program that is in conformance with Technical Specifications, regulatory requirements, commitments, and

'

industry guides or standards. The inspectors also reviewed selected maintenance work orders (MW0), preventive maintenance procedures, conducted i

i

, . ~ - -

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

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

i interviews with various maintenance personnel, and conducted tours of the plant to evaluate housekeeping practice The following aspects of the maintenance program were verified:

-

Written procedures were established for initiating requests for routine

! and emergency maintenanc Criteria and responsibilities for development, review and approval of maintenance requests were establishe Criteria and responsibilities that form the basis for designating the activity as safety or non-safety-related were establishe Criteria and responsibilities were designated for performing work ,

inspection of maintenance activitie t

- Administrative controls for special processes were establishe Methods and responsibilities for equipment control were clearly defined and establishe Written procedures were established and responsibilities designated for cleanliness control of safety-related components and system ; - Administrative controls and responsibilities for general housekeeping were establishe Methods and responsibilities have been designated for performing functional testing of structures, systems or components following maintenanc Maintenance planning and scheduling is controlled by the Work Planning Group (WPG). The WPG is comprised of representatives from the Maintenance Department, Operations Department, Engineering Department, the Quality

. Control Section, and the Health Physics and Chemistry Department. The

!

responsibilities of the WPG as defined in 50 AC-MNT-001 are: Planning, scheduling, and tracking maintenance activities Generating work order packages Reviewing the work order packages l Generating and distributing schedules of maintenance Providing MWO data for entry into Nuclear Plant Reliability Data System (NPRDS) and other management information systems i Reporting the status of maintenance and modification activities on a periodic basis l

! _ _ - -

_ _ _ - _ _ _ _ - . - _ . . _ . - - - _ . _ .

- . - _

. .- - - .

9 Ensuring that required functional tests and inspections are adequate Ensuring that work descriptions and other documentation are adequate The WPG schedules work on a weekly basi Schedules are distributed to

~

responsible supervisors on the Friday preceding the scheduled work wee All maintenance activities are tracked through a computerized system known as the Nuclear Plant Management Information System (NPMIS). All the information contained on an MWO is entered into the NPMIS. The licensee has instituted a program to generate MW0s for repetitive tasks using the NPMI This will allow the WPG to assume the function of scheduling the preventive maintenance program through the NPMIS. Previously preventive maintenance was scheduled by Regulatory Compliance, but at the time of the inspection,

' the WPG was using both NPMIS and the Regulatory Compliance generated pre-ventive maintenance schedules to track the preventive maintenance progra The inspector conducted a review of the licensee's implemented Preventive and Predictive Maintenance Program The licensee is presently in a

, Preventive Maintenance Procedure Upgrade Progra As of this inspection, the licensee had completed the review of all systems safety-related or important to operation, and determined the PM requirements for each compo-nent in the system. The licensee is presently in the process of verifying or incorporating these requirements into the PM procedures. Discussions with the licensee determined that approximately 30 percent of the PM procedures had been completed. During the review of the licensee's PM program, the inspector identified 34 PMs that were overdue as of July 28, 1986. A large portion of these PMs were overdue due to inaccessibility of the areas or to the plant conditions that presently exist. These overdue PM's are presently being tracked by the WP The licensee has instituted a program of predictive maintenance. This program consists of vibration analysis for pumps and turbines, oil analysis and infrared analysis. This data is used to trend the performance of each applicable component and is intended to reduce premature equipment failure and minimize unscheduled downtim The program is expanding in the number of equipment under evaluatio However, not all critical equipment is included, as evidenced by the recent diesel generator bearing failure. Some equipment requires some small modifications in order to draw representative oil samples. The diesel generator bearing is an example, and appears to be focusing management attention on the value of these programs and the necessity of including all critical equipment. The Predictive Maintenance Program is also used to defer performance of scheduled PMs based on satis-factory vibration analysis results and oil analysis results. Inspection Report 85-19 identified an inspector concern that the licensee had no administrative procedure that defined the Preventive Maintenance or Predic-tive Maintenance Program The inspector reviewed the administrative procedures associated with the Preventive and Predictive Maintenance

! Programs. Based on this review, the procedures do not appear to define the

Preventive and Predictive Maintenance Programs in sufficient depth.

l

. - -_

The purpose of The inspectors conducted plant tours of both Units 1 and 2.

these tours was to assess the licenesee's housekeeping practices and to observe maintenance in progres The cleanliness of both units appeared satisfactor The inspectors were unable to find any safety-related maintenance in progress during these plant tours. Observation of mainte-nance activities will be inspected during a followup inspection.

The inspectors reviewed the Unit 2 clearance index to verify that the index was being maintained in accordance with procedure 30AC-0PS-001-0S, Equipment Clearance and Tagging. The inspectors also reviewed the Unit 2 control room MWO lo This review identified that the high pressure coolant injection (HPCI) inboard isolation valve 2E41-F002 had been electrically backseated on July 26, 1986, to stop excessive packing leakage. Electrical backseating of motor operated valves is accomplished at Hatch by bypassing the open limit switch and then driving the valve disc on the backseat until Discussions the locked with Operations, rotor current of the motor is approache Maintenance and Engineering personnel determined that electrical backseating of certain containment isolation valves is performed on a routine basi These valves are:

HPCI steam supply inboard and outboard isolation valves Reactor Core Isolation Cooling (RCIC) steam supply inboard and outboard isolation valves Reactor Water Cleanup (RWCU) suction supply inboard and outboard isolation valves Recirculating pump discharge valves Review of maintenance history files determined the following:

2E51-F007, RCIC steam supply inboard isolation, has been electrically backseated at least 17 times since July 1984, the last time the valve was disassembled for maintenanc E51-F008, RCIC steam supply outboard isolation, has been electrically backseated at least 18 times since March 1984, the last time the valve was disassembled for maintenanc E41-F002, HPCI steam supply inboard isolation, has been electrically backseated at least 10 times since February 198 As a result of a violation issued in NRC Inspection Report Nos. 50-321, 50-366/85-34 concerning valve 1G31-F001 exceeding its maximum closure time of 30 seconds due to being electrically backseated, the licensee has required an engineering evaluation be performed as part of the The backseating licensee procedure, and that an MWO be generated to repair the valve. These actions has also taken actions to prevent recurring packing leak include instituting a program to replace the packing on these valves on a regular interval and also plans to live-load the packing packing leakag The to help pr backseating containment isolation valves is an acceptable practic licensee presently has a request for engineering assistance initiated to

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

'

provide guidelines for electrically backseating motor operated valves. The inspectors reviewed the electrical backseating procedure 52GM-MEL-02 If

electrical backseating is determined to be acceptable, the inspectors do not believe that the procedure is adequately as written. As written, the

'

procedure requires opening the valve from the full closed position using the temporary bypass around the open limit switch. The inspectors believe that

.! the valve should initially be opened using the control room switch and then

, eased onto its backseat using the temporary jumper around the open limit j switch. The procedure does not require that continuity and megger checks be performed and documente The procedure also does not require that an MWO be written to inspect the internals of the valve for damage at the earliest

! opportunity. As a result of this inspection effort, the inspectors deter-

,

mined that the licensee tests the stroke time of valves from light to light i as opposed to switch to light. ASME Section XI requires that valves be j timed from initiation of the actuating signal to the end of the actuating

, cycl The licensee indicated that relief had been requested from NRR to I allow timing from light to light vice switch to light, and that stroke time

testing .had always been performed from light to light at plant Hatch. No i documentation was provided by the licensee that documented relief from the i

requirements of ASME Section XI.

f In a subsequent phone call to Georgia Power Licensing, the inspectors

. requested copies of any documentation that requested relief from ASME l Section XI. The inspectors do not consider that light to light stroke time

,

testing of valves is acceptable since this method does not take into account i the time it takes the motor operator to start, the hammer blow effect

! associated with the operator, and the time it takes the valve to travel to

!

the limit switch positio Technical Specifications require that certain

valves meet finite stroke time requirements. The inspectors feel that the

'

method of stroke time testing employed at plant Hatch is inadequate. Plant

'

Hatch also has no method of ensuring that all electrically backseated valves i are timed from switch actuation during routine surveillance testing i conducted after the: valve has been electrically backseated. This item was

! identified to the licensee and will be reinspected in a later inspectio Electrical backseating of valves and light to light stroke time testing will be identified as an unresolved item (321,366/86-22-01).

j Review of Maintenance Procedures and Practices for Motor Operated Valves t (M0Vs)

j

! The inspectors performed a detailed review of the licensee's procedures for maintenance on the safety related M0Vs installed throughout the plant. The review consisted of ensuring the requirements or guidelines from the vendor technical manuals (Limitorque), IE Notices and l Bulletins, and general industry practices were contained in the

,

licensee's procedures. The following procedures were reviewed:

i

52CM-MME-005-05, Rev. O, Limitorque Valve Operator Models SMB-0 through SMB-4, Mechanical Maintenance i *

52CM-MME-006-0S, Rev. 0, Limitorque Valve Operator SMB-000 and SMB-00, Mechanical Maintenance i-

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

52CM-MME-007-0S, Rev. 0, Limitorque Valve Operator Model SMB-5 and SMB-5T, Mechanical Maintenance

52CM-MME-008-0S, Rev. 0, Limitorque Valve Operator Model SB-00, Mechanical Maintenance 52CM-MME-009-OS, REv. 0, Limitorque Valve Operator Model SB-0 through SB-4, Mechanical Maintenance 52 CM-MME-010-05, Rev. O Limitorque Manual Unit Models H0BC through H7BC, Mechanical Maintenance 52CM-MME-016-0S, Limitorque Valve Operator Type SMC-03, Mechanical Maintenance 52PM-MNT-005-0S, Rev. 3, Limitorque Valve Operator Inspection 52 GM-MNT-016-0S, Rev. O, Limitorque Valve Operator Trouble-shooting 52GM-MEL-022-0S, Rev. 1, Limitorque Valve Operator Electrical Maintenance 42SP-060986-P2-1-05, Rev. O, Inspecting Limitorque MOV Operators for EQ Compliance The inspectors considered the licensee's procedures for MOVs, in general, to be very good. The procedures flowed in logical step by step manner and were normally easy to follow. Steps contained in the vendor manuals that lacked clarity were written with more detail in procedures to prevent errors during repair. In addition, the

,

procedures were reviewed to ensure the following EQ requirements were i met:

-

The proper grease is specified for use in the actuator and limit switch gear box. All procedures specified the use of Exxon Nebula EP-0 or EP-1 for use in the actuator gear box and Mobil 28 for use in the limit switch gear box. These greases are required for use inside containment only, however, respoasible licensee personnel

! stated that all Limitorque valves would be worked utilizing the procedures reviewed and thus, these particular greases would

' be used in any valve repaired regardless of location or safety significanc Procedures required the use of the proper internal wiring between i terminal blocks, torque switch and limit switch. Any wires found

!

that could not be positively identified were required to be replaced.

i

_ ._. . ,

.

l

,

.

-

Procedures required the proper torque switches and limit switche '

Switches inside containment were required to be dark-brown or off-white while those located outside containment could be dark-brown, off-white, or re Missing T-drains or incorrectly installed T-drains on inside containment actuators are required to be installed correctly at the lowest point on the motor to ensure moisture does not short out the moto The following procedural deficiencies were discussed with responsible licensee engineers. Although there appear to be no specific regulatory requirements concerning these items, they felt the concerns were valid and will consider implementation in future revisions to the procedure The procedures did not

-

52CM-MME-006-0S and 52CM-MME-008-0S:

contain specific instructions to identify that the tripper fingers on the clutch tripper assembly should not be the same lengt Replacement fingers obtained through the supply system may be the same length and must be shortened by grinding or filin GM-MEL-022-0S: The procedure should provide a more positive means to ensure the motor pinion gear is properly installed or

,' reinstalled. The position of the motor pinion gear is critica The pinion must be mounted with the gear hub facing the motor on SMB-0 actuators. On SMB-1 through 4 actuators, the hub must face opposite the motor. Installation in the incorrect position mayAs cause gear tooth failure due to insufficient tooth engagemen presently written, the procedure only requires the installed position of the pinion be note GM-MEL-022-0S: The procedure should contain instructions to I -

ensure the torque switch is balanced during installation.

l l

j

-

52CM-MME-007-0S, As previously stated the maintenance procedures were well written, however, a review of this procedure indicated it was basically a word for word copy from the vendor technical manua The other procedures contain numerous additions and j clarifying instructions not contained in the technical manuals.

l This is due to increased experience from valve actuator mainte-nance. The licensee stated the reason this procedure was lacking compared to the others is that there are very few SMB-5 and SMB-5T actuators installed in the plant. Thus, the experienced gained from working the large number of the other types of actuators was not availuble for these actuator types. The licensee stated that as more maintenance experience was gained it would be incorporated into the procedure, as applicabl ._ _ _ _ _ _ _

_ _ _ _ _ _ _ _ _ _ _ _ _ _ _

a

.

b. Discussions with licensee engineers and maintenance personnel, and a review of internal correspondence, indicated that Hatch has had several problems associated with Limitorque operators. Plant Hatch has approximately 650 Limitorque motor operated valves, of which 220 are safety related. The problems discussed included the following:

-

Inspections at Plant Hatch revealed deficient grease in the main gear housing, the limit switch gear box, and/or the spring pack in 28.6 percent of the actuators inspected to date. In addition, separated grease was found in the majority of the actuators inspected in the warehouse. These problems are due to mixing of greases, harsh environments, or the tendency of grease to separate over long periods of tim To resolve and prevent grease problems which could cause improper Limitorque operation, several steps have been taken at Plant Hatch:

(1) Identification of actuators with mixed greases and disassembly, cleaning, and relubrication with Exxon Nubela EP- (2) A memo was sent to Maintenance personnel, the lubrication guide was changed, and procedures were changed to note the use of Exxon Nebula EP-0 or EP-1 in Limitorque actuator (3) Lubrication analysis procedure 50PM-MON-002-0S was writte In addition to analyzing the oil, this procedure may be used

to sample and analyze the condition of grease without disassembly of the Limitorque actuator. Use of this method

,

will provide more confidence in the condition of grease in i Limitorque actuators.

l l (4) Change over to Mobil Grease #28 for use in limit-switch gearboxes for higher temperature resistanc Several Plant Hatch actuators have also been found without torque switch limiter plate However, in lieu of relying on limiter plates to ensure the torque switch setting does not exceed the maximum value recommended, new Limitorque corrective maintenance

.

procedures and the new set-up and test procedure will require l listings of "as found" and "as left" settings with notes to not exceed the maximum settin In addition, training procedures for setting Limitorque torque and limit switches have been evaluated to ensure corrective methods are use Of all the Limitorque actuators inspected at Plant Hatch, 2 l percent had deficient limit switches and/or torque switches (not

including grease related problems). In 7 cases limit switch l

rotors and/or baseplates were found cracked or broke Replace-

-

ment of deficient switches is being done as they are identified

!

l l

l

!

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

in inspection New procurement policies and maintenance proce-dures will ensure that the correct switches (material type) are use Several operators have been found without the drilled drain plugs (T-drains) installed as per Limitorque instruction As a result, a memo will be sent to all Maintenance personnel concerning the installation of T-drains in Limitorque motors with

"RH" insulation. In addition, all actuators with RH insulated motors will be inspecte T-drains will be installed on any motors without the drains installed according to Limitorque's instruction All gaskets are replaced during reassembly of the operators after inspection. In only one case have the gaskets been found damage In one other case a gasket with the wrong thickness size was found as the cause of mechanical problems with the actuato Another problem has also been identified in Plant Hatch inspec-tions to date. Of the Limitorque operators inspected, 34.7 percent have been found to have problems with the declutch to manual operation " system". These problems were caused by poor operating knowledge, i.e., trying to force the declutch lever to put the operator in the manual mode. Bent or twisted declutch shafts were found in 22.5 percent of the operator To reduce these type problems, directives were sent to both Maintenance and Operations personnel instructing them on the proper methods of putting a Limitorque into " hand" operation. Also, training procedures have been evaluated to ensure proper instruction is give The licensee's program and procedures for maintenance, inspection, and problem resolution for M0Vs appeared to be adequate. It should be noted that implementation of the program and actual adherence to the procedures was not evaluated during the course of this inspection, due to lack of maintenance in progress. These items will be addressed in a future inspection when actual maintenance in progress can be observed.

9. Maintenance Implementation (62700)

The inspectors reviewed selected maintenance work order packages (MWO),

procedures, maintenance history files and interviewed several maintenance supervisor The maintenance work associated with Licensee Event Report (LER)86-027, Valve Leakage Found During Local Leak Rate Test, Required Rector Shutdown, was reviewed. On June 12, 1986, Hatch Unit I was required to shut down due to the failure of drywell vent valves, IT48-F319 and IT48-F320, to pass the required local leak rate tes Maintenance work orders (MWO) 1-86-5083 and 1-86-5084 were written to repair 1T48-F-319 and IT48-F-320 respectivel The inspector noted that MW0 1-86-5083 and MW01-86-5084 indicated in block 27 of the work order continuation sheets that the valves 1T48-F-319 and

1T48-F-320 were repaired and replaced in the system on June 14, 1986. On June 15,1986, both valves were again removed from the system. The MW0s did not state why the valves were removed from the system, or why they were being reworked. The licensee informed the inspector that the valves were removed from the system for rework because they failed a preliminary leak rate test. This test was not documented in the MWO package. Administrative control procedure 50AC-MNT-001-0S, Maintenance Program, Step 8.6.2 requests that the results of all inspections and tests performed shall be documented and further action based on results and the documentation will become part of the MWO package. The licensee failed to document the results of the preliminary test associated with MW0s 1-86-5083 and 1-86-5084, concerning repairs on drywell vent valves 1T48-F319 and IT48-F320. The licensee was informed that the above failure to complete the MW0 per established proce-dures was a potential violation (321/86-22-02). Additionally, a review of MWO 1-86-5084 indicated the following deficiencies:

- New i inch copper tubing with new nuts and ferrels were installed on valve IT48-F319 operator with out a stock material issued (SMI) form being completed and Quality Control did not perform a material veriff-cation. Maintenance administrative procedure 50AC-MNT-001-05, revision 4, dated November 27, 1985, requires that if material is needed during MWO activities, an SMI should be completed in accordance with material procedures and Quality Control will perform material verification activities as part of the QC inspection program. The licensee was informed that this was another example of potential violation (321/86-22-02). Other examples of failure to adequately document and verify the material used includes MWO 1-86-0389, replacement of resilient seats in valve 1T48-F320, where Swageloc brand fittings and i inch copper tubing from the maintenance shop stock were used to make repairs to valve IT48-F320. The work description for MWO 1-84-8358, repair of valve 1T48-F319, states: removed operator and found one key was missing and the other key was half sheared, also sleeve where key fits was wallowed ou The replacement of the keys with proper

' material was not documented or QC verifie The torque wrench used to torque the flange bolt and operator mounting bolts for valve 1T48-F319 was not documented in block 27 (actual work performed section) of the MW0. Revision 5, dated July 29,1986, of maintenance procedure 50AC-MNT-001-05 requires documenting identifica-tion numbers of the calibration tools used, and the calibration due date of each tool in block 27. The quality control (QC) inspector did

,

document the tools used on the QC inspection checklist. MWO 1-86-5083, j

'

for valve IT48-F320 also lacked calibration tool documentation in block 27 of the MW Other examples of where torque wrenches were not documented as being used are MWO 1-84-8403 and MWO 1-84-8358.

t i

The review of MWO 1-86-5083 and 1-86-5084 indicated that the valve flange bolts were torqued to 675 foot pounds using a 250 foot pound torque wrench l in combination with a torque wrench multiplier. A review of measuring and test equipment records showed that the torque wrench was controlled, calibrated and properly identifie The torque wrench multiplier had

identification number, 391A268517, but no documentation other than the operating instructions for Model 391A torque multiplier to identify the multiplier and its accuracie The Model 391A torque multiplier uses a planetary geared action to produce a 6 to 1 torque ratio with a stated accuracy of 5 percent. The multiplier has a rated input capacity of 200 ft-lbs with a rated output capacity of 1200 ft-lb CFR 50, Appendix B, Criterion X11, states: Measures shall be established to assure that tools, gages, instruments, and other measuring and testing devices used in activities affecting quality are properly controlled, calibrated, and adjusted at specified periods to maintain accuracy within-necessary limit The licensee failed to assure that the torque multiplier was properly calibrated or adjusted to maintain accuracy within necessary limits. The licensee informed the inspector that they contacted the manufacturer of the torque multiplier and the manufacturer stated that each torque multiplier is calibrated during assembly and its torque ratios and accuracy is uniquely determined and documented. The licensee stated that they did not receive this manufacturer's documentation or test the multiplier themselves to determine its accuracy. The licensee was informed that failure to ensure the torque multiplier's accuracy was a potential violation (321, 366/86-22-03).

Additionally, the use of a torque multiplier with a given torque wrench may cause the mechanic to under or over torque a bol The torque wrenchs are calibrated with an allowable accuracy of 4 percent of full scale in the clockwise direction and 6 percent of full scale in the counter clockwise direction. That would allow a 250 ft-lb torque wrench to have a 10 ft-lb error band for clockwise operation A 1000 ft-lb torque wrench would have a 140 ft-lb error band. If a mechanic torques a bolt to 675 ft-lb using a 100 ft-lb torque wrench, the bolt could be torqued to 675 ft-lb 40 ft-l If the mechanic used a 250ft-lb torque wrench in combination with a 6 to 1 torque multiplier, the bolt would be torqued to 675 ft-lb 60 ft-lb, due to the multiplication of the torque wrench (112.5 ft-lb) setting plus its maximum allowed accuracy range of 10 f t-l The accuracy of the torque multiplier is 5 percent, and if 100 ft-lb was applied, the results could be 600 30 f t-lb. The additional error of 30 ft-lb would also effect the total torque value on the bolt. The licensee informed the inspector that they would review the tolerances associated with bolt torque requirements and determine if the torque wrench accuracies in combination with torque multipliers is acceptable. The above determination will be an inspector followup item (321,366/86-22-04).

The inspector reviewed several MW0s associated with the calibration of the turbine building main steam leak detection instrumentation, IU61-N101-A through D. When performing the calibration on October 16, 1984, two instruments were found out of tolerance, IU61-N101A and B. On March 28, 1985, Unit 1 received an invalid half group I isolation on reactor protec-tion system (RPS) "A" channel due to IU61-N101A and 1U61-N101B being out of tolerance, which caused a RPS actuation signa When performing the latest calibration of these instruments, on December 4, 1985, all 1U61-N101-A through D temperature instruments were out of tolerance. The L

inspector expressed concern about the instrumentation trending, and the cause for the instrumentation drift. The instrument and control supervisors indicated that they believed the drifting was attributed to the environ-mental conditions of the turbine building when the instruments were calibrated, i.e., high temperature condition in the summer, and lower temperatures during the fall and winte The inspector requested the deficiency report or the request for engineering (RER) review on these instruments. The review of the deficiency report or RER will be an inspector followup item (321/86-22-05).

10. Corrective Action (92720)

' General The inspectors performed an inspection of corrective action in the maintenance area to assess the status of this functional area, and to determine whether the licensee had developed a comprehensive corrective action program to identify, follow, and correct safety related problems. The following criteria were used during the review to assess the overall acceptability of the established program:

- Had the licensee established policies and administrative controls for identifying problem Had Management controls been established for the tracking and resolution of problems identified by:

Operational events Internally identified problems Quality Assurance Audits

>

NRC inspection findings Employee concern programs and concerns brought by external persons or organizations The documents listed below were reviewed to determine if these criteria had been incorporated into the corrective action program.

! 10AC-MGR-04-0, Deficiency Control System Revision 0 10AC-MGR-005-0S, Corrective Action Trending and Tracking, Revision 1 AG-MGR-02-1284, Quality Concern Program Revision 1 AG-REG-04-185N, NRC/QA Item Corrective Action Program, Revision 0 DI-REG-07-1085N, NRC/QA/INP0 Item Corrective Action Program, Revision 1 43RC-CPL-003-0S, Defects and Noncompliance 0

.-. _

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

QA-05-01, Field Audits, Revision 15 QA-05-02, Corporate / Supplier Audits, Revision 14 QA-05-13, NRC Open Items Control, Revision 13 QA-05-17, QA Surveillance, Revision 2 QA-05-20, QA Trend Program, Revision 2 DI-QCX-02-0485, Quality Feedback Report, Revision 0 During the inspectors' review of the corrective action programs, examples were found that raised concerns in the areas of:

Failure of corrective actions to prevent reoccurrence The failure of event cause analysis to determine actual root cause Too narrow scope in identifying previous similar events in LER The description of events in some LERs did not clearly describe all aspects of the even Excessive personnel error rat Continuing and repetitive reactor water cleanup (RWCU) system isolations from high room temperature and high differential flow over a two-year period are an example of inadequate corrective actions. The majority of these engineered safeguards feature (ESF) equipment actuations were caused by high RWCU room temperatur Several high differential flow actuations, however, resulted during valving operations for the RWCU filter /demineralizers. These isolations appear to have been caused by operating the valves to fast, and a failure to follow procedures. The licensees root cause analysis and corrective actions appeared to be less than timely for correcting the RWCU system isolations caused by high room temperatures. A new ATTS system was installed in 1986 which lowered the room temperature trip setpoint from

'

145 F to 124 F, and physically mounted the new temperature sensors in the overhead of the room. The lower temperature setpoint established by this modification resulted in an increase in ESF actuations (approximately 12 actuations over a 3 month period). The licensee stated that the trip setpoint was lowered to this value because adequate data was not provided to General Electric (GE) for analysis prior to the installation. This data has subsequently been provided to

,

GE and their analysis will permit the room temperature setpoint to be raised to 150 F which should resolve this problem. The concern is that the measures taken during this 2 year period did not promptly correct the identified problem which resulted in excessive challenges to the ESF equipmen Another example of event analysis deficiencies was the

assumptions made without technical basis in LER 85-043, Failure of Containment Penetrations to Pass Local Leak Rate Tests. The statement was made in the analysis section of this LER that the degradation of the valves to the point that they could not pass the leakage tests u probably occurred toward the end of the surveillance interval. Subse-quent failure of valves identified in this LER required a plant shut-down to repair the excessive leakage. Licensee management stated during interviews that they eventually performed additional analysis for the valves identified in LER-043 to better support the analysis statement. A significant number of events reported in accordance with 10 CFR 50.72 and 50.73 that were reviewed had the cause of event listed as unknow Although the exact cause of events cannot always be identified, the proportionate number of unknowns appeared excessiv Additionally, a number of reportable events reviewed did not list previous related events as similar. An example were isolations of the RWCU system which occurred on December 21, 1985, and January 10, and 15, 1986. All of these events identified the cause as personnel error, two I&C technician errors and one system alignment error. Although each of these events involved inadvertent RWCU isolations by I&C personnel, and in two of the events improper use of jumpers, each report indicated no previous similar even It appears that the licensee classification of similar events may be too narrow in scope to ensure that trends are identified and corrective actions taken to prevent recurrenc The LERs and event reports reviewed, indicated a relatively significant amount of personnel errors are continuing to occur at Plant Hatch. Although the inspectors realize that totally eliminating personnel errors is virtually I impossible, many of the personnel errors reviewed indicated a potential for reduction or elimination. The corrective actions necessary to reduce the error rates might include many different directions including procedure revisions, procedural adherence, improvements in

'

labeling and human factors areas, independent verification, additional training, and system design changes. Many of the new programs being i established by the licensee have the potential to meet these objectives and to significantly impact the personnel error rates, providing proper emphasis is placed on identifying all contributing causes and trending l similar events. At the time of the inspection, however, these programs were in the early stages of implementation and had not produced a notice-able effect on Hatch personnel error rates. This area will be reviewed again during the maintenance followup inspection to be conducted, b. Deficiency Reporting System The inspectors determined that licensee management had established various administrative control systems to identify and correct conditions adverse to quality. The administrative controls employed l for identifying and resolving defects are as follows:

A maintenance work order (MW0) is used to control the correction of defects in station equipmen .

.

A request for engineering review (RER) is used to initiate and control modifications to station equipmen The material deficiency section of a Deficiency Report (DR) is used to control and process non-conforming materials, parts, and components identified during receipt inspection A temporary procedure change form or a procedure / instruction request form is used to control the development of or revision to Plant Hatch procedures and instruction The training request or training directive is used to identify, request, and document the training needed when a problem was caused by personnel erro In addition to the administrative controls described above, a deficiency report (DR) is prepared for deficiencies which have been classified as significan This ensures that appropriate management review and root cause analysis are perforited for identified deficiencies. Guidance in the classification of significant conditions adverse to quality has been delineated in administrative procedure 10AC-MGR-04-0. Organizational responsibilities for ensuring effective implementation of the Deficiency Control System has also been delineated in writin c. Quality Trend Programs Licensee management has established a Corrective Action Trending and Tracking Program. This program ensures that problems identified under the deficiency reporting system are assessed for recurrence, in addition to providing a formal process for assessing and tracking problems identified by the NRC, INP0, the Safety Review Board (SRB) or Quality Assurance Organization. Administrative procedure 10AC-MGR-005-0S delineates the methods and assigns responsibilities for:

Trending material / operational problems to identify, correct, and report program, operational and generic material deficiencie Assessing and correcting deficiencies identified by outside agencies or corporate inspection and audit program Assessing and correcting generic industry problems identified throughout the nuclear industry and which are considered applicable to Plant Hatc .- - __ _

.

.

The scope of the trend program includes monitoring deficiencies on any system, structure, or component whose failure was the cause of a condition to be reported by a licensee event report (LER), or by the reporting requirements of 10 CFR 2 Responsibility has been assigned to the Regulatory Compliance Department for monitoring DRs that identify non-conforming material / operational conditions which have been determined to be reportable in accordance with the requirements of administrative procedure 10 PC-MGR-04- Nonreportable DR's are also trended by Regulatory Complianc The Corrective Action Trending and Tracking Program provides for the performance of a monthly trend sort to evaluate potential trend Adverse trends are reported to the superintendent QC and the cognizant department manager or his designee. The program requires that a DR be prepared by the cognizant department manager to ensure evaluation and root cause analysis by plant managemen The inspectors reviewed the following documents in connection with the assessment of the corrective action trending and tracking program:

NRC/QA item trend monthly reports for the period June 1985 to January 198 NRC/QA item trend monthly reports for the period December 1985 to June 1985 The inspectors concluded that the Corrective Action Trending and Tracking Program appears adequate. Implementation of the program is less than fully effective however, based on the following observations:

  • -

Increase in backlog items from 35% in May to 60% in June as described in report dated July 8, 198 .

A total of 14 items requiring manageaent's attention over the one f

year time span reviewed.

! Missed milestone dates are attributed by the licensee as the cause for the above. The arrangement for Regulatory Compliance to provide

, department managers with a " Weekly Overdue Item Status Report" will hopefully correct this lack of adherence to established milestone dates for prompt correction of identified deficiencie Regulatory Compliance has been assigned responsibility for the detailed implementation of the corrective action program through all phases of receiving, issuing, and processing NRC/INP0/QA items. The administra-tive control for implementation of the corrective action program l delineated in administrative guideline AG-REG-04-1085N and department l

instruction DI-REG-07-1085N were reviewed by the inspectors and found

'

to meet regulatory requirements and licensee commitments.

i

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

.

d. Quality Assurance Audits Technical Specification (TS) Section 6.5.2.8 delineates the require-ments for the conduct of audits under the cognizance of the Safety Review Board (SRB). The inspectors reviewed the following documents and verified that audits of corrective actions specified in TS section 6.5.2.8.2.c are conducted at the required frequency:

Plant Hatch QA Department, Annual Audit Planning Matrix, Schedule and Status Report for the year 1985, dated 12/5/85 Plant Hatch QA Department, Annual Audit Planning Matrix, schedule and status for the year 1986, dated 5/13/8 The following audits were reviewed by the inspectors to verify compliance with program procedures and regulatory requirement Audit Number: 85-CA-1 Date Audited: March 18,1985 - April 11,1985 Activity: QA Audit of Corrective Action Program Audit Number: 85-CA-2 Date Audited: September 17, 1985 - October 25, 1985 Activity: QA Audit of Corrective Action Program Audit Number: 85-QC-2 Date Audited: June 10-20, 1985 Activity: QA Audit of Quality Control Program Audit Number: 85-QC-3 Date Audited: September 3-16, 1985 Activity: QA Audit of Quality Control Program Audit Number: 85-QCA-1 Date Audited: February 18-28, 1985 Activity: The QC Section and other QC programs being implemented by other activities. An overall assessment of plant activities based on the corporate QA program, Project QC Section, and other related QC activities being implemented by activities other than the Quality Control Sectio Audit Number: 85-MA-1 Date Audited: February 11-25, 1985 Activity: Quality Assurance Audit of Maintenance Audit Number: 85-MC-2 Date Audited: June 11-28, 1985 Activity: Quality Assurance Audit of Material Controls

_ _ _ _ _ . _ . _ _ _

. _ ._. ..

.

'

Audit Number: 86-CA-1 Date Audited: March 3,1986, April 1-4,1986, April 7-8,1986, April 11, 1986, April 21, 1986 l

Pursuant to the review of the above audit reports and the corrective actions initiated for identified deficiencies, the inspectors concluded that the audit program and its implementation appears to be adequat The inspectors determined that licensee management had established a QA trend program. The objective of this program is to provide appropriate management with advanced warnings of real or potential problems as indicated by possible adverse trends. Problems identified through this process are then further evaluated by assessments, audits, surveillance or investigatio The data base of the QA Trend Program consists of the results of all QA audits, QA surveillance findings (except housekeeping), NRC findings, and INP0 evaluations. The data base sample covers three months time interval All data collected is classified in accordance with the following code:

0A TREND CODES Principal Trend Code Description E Personnel Error C Construction / Installation D Design j Q Programmatic The principal trend codes are to be applied based on the following definitions:

E - Personnel Error - An error made when a written procedure was not followed or because personnel did not perform in accordance

with accepted or approved practic C - Construction / Installation - This classification is assigned to problems reasonably attributed to the construction or installation

. of a system, component, or structure. For example, failure to comply with approved drawings and/or designs which result in an

"as-built" condition which deviates from the intended conditio This may be shown by physical inspections showing deviations or by the failure of the component to meet its functional test require-

, ments.

!

i

._- . - . _ _ .______ . _ _ _. . _ _ . . _ _ _ _ _ _ _ . _ _ _ _ _ _ _ . . _ _ _ _ . _ _ _ . . _ . _ _ ,.. _ ____ __ _ ,, _ .

0 - Design - This classification is assigned when the problem is reasonably attributed to design of a system, component, or structure. For example, problems that are traced to such things as incompatibility of materials, loss of design control, or choice of components which fail to meet specified functional requirements or performance specification Programmatic - This classification is assigned to problems caused by the failure of management or management systems (e.g.,

breakdown in administrative controls, preventative maintenance program, surveillance program, or quality assurance controls).

This would include FSAR commitments not met or implemented and license conditions not implemented.

Each principal trend category is further sut' divided into cause code sub-categories.

A complete graph of each principal trend code is then prepared using the guidance delineated in procedure QA-05-20. Additionally, reviews and verification of potential problems as depicted by the graphs are performed in accordance with the administrative controls described in this procedure.

The following documents were reviewed by the inspectors in connection with review and assessment of the QA trend progra Interoffice correspondence from P. E. Fornel-QA Site Manager to D. S. Read - General Manager RE: NRC/QA Item Trend Analysis, First quarter 1986, dated March 4,198 Interoffice correspondence f rom L. C. Byrnes to D. S. Read, '

Subject: Supplement to the First Quarter 1986 Trends Report, Interoffice Correspondence from Q. M. Fraser - Acting Site QA Manager to D. S. Read-General Manager Quality Assurance, RE:

NRC/QA Item Trend Analysis Second Quarter 1986, dated June 6, 1986.

The Second Quarter Trend Report identified an increased number of personnel errors caused by a lack of attention to detail or failure to follow procedure. These personnel error findings were identified in the areas of design change requests, emergency plan, plant operations, fire protection, measurement and test equipment and corrective actions. In discussions with licensee management, the inspectors were informed that the increase in personnel error can be attributed to fatigue and preoccupation with Unit 1 outage activities, and the time constraints associated with completing the outag . - - - . . . - _ _ . -

.

l,

!

i The inspectors were further told that the majority of personnel errors l were made by contractor personnel. In response to this finding, licensee management has instituted administrative controls which require vendor activities onsite to be in accordance with all Plant Hatch procedures and the QC program. A selected list of procedures requiring the attention of vendor / contractor personnel prior to any

,

onsite activities have been prepared and is included as an Appendix to

'

all requests for procurement of service The inspectors concluded that the licensee's QA Trend Program meets the requirements of licensee canmitments delineated in ANSI 18.7-1976, and appears to be effectively implemented. Personnel errors identified may be indicative of a generic problem of loss of control of personnel / procedural interface requirements within the " work control" process. This assessment is substantiated by the fact that numerous new procedures are presently being prepared by the Procedures Upgrade Program (PUP). Additionally, the requirement to ensure that plant f personnel have been trained to the new procedures have not yet been fully accomplished. The inspectors commented favorably however, upon

licensee management arrangements to have plant personnel participate in
the validation of the new procedure NRC Inspection Findings l The inspectors reviewed licensee program documents and conducted interviews with licensee management to assess the status of NRC open

,

items. The following document was reviewed in connection with this effort:

Interoffice correspondence LR-QAM-004-0786 from 0. M. Fraser,

Acting QA Site Manager, RE: Summary of Plant Hatch NRC/QA Status

dated July 2, 198 The inspectors determined that for the period June 1 through June 30, i 1986, two NRC inspection reports were received by the QA department

, that resulted in eight items being opened and two items being closed.

l Also, of the present 100 NRC open items, 67 are awaiting NRC review

after completion of corrective action and 33 have made satisfactory

progress for the mont The inspectors concluded that adequate corrective actions are being

,

implemented by licensee management regarding NRC inspection finding Internally Identified Problems The inspectors reviewed the following documents to assess the status of

.

licensee internally identified problems documented as Deficiency l Reports (DR):

.

Interoffice Correspondence No. LR-ENG-006-0786, from T. R. Powers t RE: DR Report No. 86-2

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

_ _ _ . _ _ _ _ _ _ _

.

.

Interoffice Correspondence No. LR-MGR-058-1085, from H. C. Nix RE:

Deficiency Report Trend Analysis Report (No. 85-6)

Interoffice Correspondence No. LR-MGR-102-0885, from H. C. Nix RE:

Deficiency Report Trend Analysis Report (No. 85-5)

Deficiency Report No. 86-2 indicated that through April 30, 1986, the number of DRs was 93 This showed an increase of 352 DRs over the number of DR's generated for a similar period in 1985. Licensee management attributes this increased rate of DRs to the Unit 1 outage (November through May), when more DRs were generated because of work activities associated with plant shutdow The breakdown of the DR's identified in DR Report No. 86-2 are as follows:

219 were due to component failure 214 were due to design, manufacturing, construction 249 were due to personnel error 118 were due to defective procedure 5 were due to external causes 94 were due to other cause Of these, 68 were found not to be valid deficiencies Note that 35 DRs had inadequate responses on which to base cause code assignments. These responses should be upgraded before the next report is issued.

I

! Figure 13A, Personnel Errors by Responsible Department, identified the

! three largest contributors as 29.9 percent unknown; 25.2 percent due to maintenance; and 18.9 percent due to contractors. The large percentage (29.9 percent) identified as unknown may be indicative of inadequacies in licensee's root cause analysis of personnel error Figure 14A, Personnel Errors by Cause Code, identified the three largest contributors as 42.6 percent due to poor craftmanship; 2 percent as failure to follow procedures; and 14.3 percent as failure l to use reference materia The inspectors concluded that the information provided by the DR trend

'

analysis is similar to that obtained from the OA trend program. The DR trend analysis further corroborates the inspector's previous assessment of a generic problem involving a breakdown of the controls associated with the personnel / procedural interface requirements within the work l

_

.

.

control proces Corrective actions that are teing implemented by licensee management include additional training of plant and contractor personnel, and a Procedures Upgrade Program which is presently ongoing. These two issues are discussed further in other parts of this repor g. Operational Events The inspectors reviewed the following documents to assess the status of licensee operating events that are identified as licensee event reports

'

(LER):

Interoffice Correspondence No. LR-ENG-026-0286 from T. R. Powers, RE: LER Trend Report No. 86-1 Interoffice Correspondence No. LR-MGR-004-0186, from C. T. Jones, RE: LER Trend Report No. 85-4 Interoffice Correspondence No. LR-MGR-024-1258, from C. T. Jones, RE: LER Trend Report No. 85-3 Interoffice Correspondence No. LR-MGR-095-1185, from D. T. Jones, RE: LER Trend Report No. 85-2 Licensee LER Trend Reports were developed by evaluating each LER and i assigning it to one of the following cause codes:

l 1 A Personnel error

"

B Design, manufacturing, construction C External cause D Defective procedure E Component failure X Other Each cause code was then further broken down into areas of signifi-cance, to identify, as far as possible, the reasons for component failure, personnel error, etc., root cause analysis may not be

,

performed to ascertain any single root cause for all reported problems.

< Depending on the nature of a problem and the availability of manpower

. and equipmen LER Report No. 86-1 trended events which occurred from January 1984 through December 1985. Significant parts and trends identified in this

report are as follows. There was an increase in the rate of LER

' occurrence in that the monthly rate of LER's for 1985 was 71/4 as compared to Si for 1984. Additionally, LER's attributed to personnel errors showed a constant trend of 32% and 33% over the time spans from November 1, 1984 through October 31, 1985, and January 1, 1985 through December 21, 1985, respectivel _ _ _ _ _ . _ - _ _ - . _ - _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ . ___

.

The inspectors determined that the data provided by the above trend report is consistent with the evaluation independently arrived at by the NRC by analysis of reported LERs. The inspectors also concluded that the data from the LER Trend Report is consistent with that provided by the QA Trend Program and the Deficiency Trend Analysis Report This is based on the observation that a generic problem involving lack of control of personnel / procedural interface require-ments within the " Work Control Process" appears to exis Component failure trends showed that 70 percent of the 21 Unit 1 LERs resulting from component failures were initiated by probicms within 5 systems. For Unit 2, 65 percent of the 19 LERs resulting from compo-nent failures were initiated by problems within three systems. Figure 16A, Lurrent Component Failure Causes, shows 32 percent of the causes as unknown. This may be indicative of inadequacies in the licensee's root cause analysis precess. Additionally, the large percentage of component failures initiated by unknown causes may also be interpreted as indicators of problems within the work control process as it per-tains to plant / personnel interface requirements. Corrective actions that are being implemented by licensee manageinent such as upgrade of personnel training, and upgrade of procedures are discussed in detail elsewhere in this repor Within this area two violations were identified and are discussed in the following paragraph Failure to Follow Procedure A sample size of 30 closed out safety related maintenance work orders (MWO) were reviewed by the inspectors to verify procedural compliance I with the QA program. Three MWO's were identified which indicated that maintenance activities were conducted in a manner such that deficien-cies found during performance of these work activities were not promptly documented, evaluated, and corrected. Maintenance Work Order No. 1-86-3998 required performing the overspeed test on the HPCI turbine. The test acceptance criteria was not achieved, and contrary to procedural requirements, maintenance was performed to correct the deficiency. Documentation and evaluation of the deficiency for root cause analysis was never performed for this identified proble Maintenance work order No. 1-84-4872 required that a turbidimeter be inspected and repaired to facilitate calibration of the instrument. It was discovered that the problem was not a defective instrument, but rather incorrect calibrating guidance contained within the calibration procedur Contrary to the QA program requirements, a deficiency report was not prepared to document and initiate corrective actions for this identified deficienc Similarly, MWO No. 1-86-3426 required that the fuel pool cooling holding pump be investigated and repaired to correct a problem with the pump failing to start when the control switch was placed in auto or manual position. Investigation revealed that the thermal overload relays had tripped and the motor full-load running current was not at the rated value. A deficiency report to

.

document and initiate root cause analysis of this problem was not prepared, and no further actions were taken. These failures to adhere to the requirements delineated in procedures 50-AC-MNT-001-05 and 10 AC-MGR-004-0 to assure that deficiencies identified during mainte-nance activities are documented, evaluated, and promptly corrected is identified as violation (50-321,366/86-22-06).

h. Operational Experience Feedback The Maintenance Manager indicated at the entrance meeting that operational experience feedback information does not go past the maintenance foremen (i.e., to individual mechanics), but is instead inserted into procedures as required. This method of dispensing operational experience may be adequate in some circumstances, but this approach does not fully meet the intent of NUREG 0737 Item 1.C.S. In some cases, it is essential that actual formal training be provided to the individuals responsible to emphasize the reasons for the changes and the safety significance of errors. This training can also be in the fonn of a discussion of the sequence of events and contributing causes, or how a particular event at another plant relates to mainte-nance at Hatch. Required reading can also be utilized in certain applications as a method of dispensing operational feedback to maintenance personnel as it is in operations. Operating experience needs also, in some cases, to be incorporated into formal training programs such as those presently being implemented at Hatch in the maintenance area. Placing significant operating experience into the training and retraining programs will ensure that new personnel can also avoid repetition of errors and events that have occurred within Hatch or at other facilities. The key to ensuring that significant operational experience is distributed adequately to all affected personnel is a formal documented program, with an adequate feedback mechanism to ensure all affected individuals have received the training or reviewed the material. The licensee received a previous violation (321,366/85-07-03) for failure to establish an operational experience feedback program for maintenance personnel to ensure that they are informed and training programs upgrade Due to an apparent continued reluctance by the licensee to establish a formal program, in addition to the informal newsletters and bulletins, this violation could not be closed by the inspector NUREG-0737, Item I.C.5 delineates the requirements for preparing and implementing procedures governing feedback of operating experience to plant staff on or before January 1,1981. Paragraph I.C.S.7 further states that periodic internal audits shall be performed to assure that the feedback program functions effectively at all levels. Engineering procedure 30AC-0PS-005, implements this requirement and requires that audits of the functioning of the operational feedback program at all levels be audited on a biannual basis. In discussions with licensee management the inspectors determined that an audit of the Operating Experience Feedback Program was conducted by the QA organization and i the findings are documented in Audit Report No. 85-TR- This audit

{

__ __

'

.

was not intended to meets the requirements for audit of the program delineated in procedure number 30 1AC-0PS-003. This audit was a small part of an overall QA audit, and addressed only the fact that there was a procedure in place for controlling operating experience feedback to Operations personne The inspectors were informed that audits of the feedback of operating experience program performed under the administrative controls This failureofto the

! Engineering Department have never been conducte conduct periodic audits as required by NUREG-0737 paragraph I.C.S.7 and Procedure 30 AC-0PS-003 is identified as violacion (50-321, 366/86-22-07).

1. Control of Contract Maintenance Personnel Due to previous allegations and problems involving contract personnel and valve maintenance, the licensee offered to make a concerted effort

.

'

to upgrade Georgia Power controls over contractors on site. The

' licensee indicated that contract personnel would now be required to provide documentation of training and qualifications related to work tasks to be performed. Previously, the contracting firms certification l of individual qualifications had been considered adequate. The contract personnel are also being required to complete at least one week of site specific training. In addition to GET training areas such as radiation protection and security, this training emphasizes This the time need is to follow procedures in performance of maintenance task also utilized to familiarize each contract maintenance individual with the procedures relating to the specific tasks that he will be preformin The licensee indicates that the number of Georgia Power personnel directly supervising the work of contract personnel will be substantially increased. The total number of contract firms A supplying review of maintenance personnel will be reduced for added contro the personnel error rates by the inspector, involving both contract and Georgia Power personnel, indicated that these additional controls were definitely warranted. The impending refueling outage should provide an indication whether these newly established contractor controls will

be effective.

!

!

!

_ -

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

Retrieved from "https://kanterella.com/w/index.php?title=IR_05000321/1986022&oldid=3234610"