ML20206P398
| ML20206P398 | |
| Person / Time | |
|---|---|
| Site: | Beaver Valley |
| Issue date: | 08/14/1986 |
| From: | Carey J DUQUESNE LIGHT CO. |
| To: | Murley T NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION I) |
| References | |
| 2-NRC-6-089, 2-NRC-6-89, NUDOCS 8608270199 | |
| Download: ML20206P398 (7) | |
Text
.
gVg
'A@
M 2NRC-6-089 Beaver Valley No. 2 Unit Project Organization Telecopy 42 5
Ext.160 f,h.so O2$"
August 14, 1986 Shippingport, PA 15077 United States Nuclear Regulatory Commission Region I 631 Park Avenue King of Prussia, PA 19406 ATTENTION:
Dr. Thomas E. Murley Administrator
SUBJECT:
Beaver Valley Power Station - Unit No. 2 Docket No. 50-412 Systematic Assessment of Licensee Performance (5 ALP)
Report No. 50-412/85-98 Gentlemen:
Thank you for your treeting with us on July 9, 1986 to discuss the Beaver Valley Power Station Unit No. 2 SALP Report for the period of. April 1, 1985 through March 31, 1986.
In addition to the items discussed at the July 9th meeting, the following items summarize the actions taken or planned to be taken to address the concerns which were identified in your letter of July 2, 1986.
F.nctional Area No. 4, Support Systems and Functional Area No. 9, Assurance of Quality The incident of Inspector error is described in these functional areas.
The project has always been concerned with the potential of inspector error; but since the incident reported during the previous SALP period (where we initiated a 10CFR50.55(e) report), we have increased our attention to the re-verification and sampling of inspectors' work.
Formal summaries of our results have been transmitted to Senior management for their review and information.
We have re-inspected 7.,601 attributes previously accepted by QC Inspectors.
This re-ir.spection resulted in 7,511 acceptable conditions and 90 unsatisfactory conditions. For information,1,832 of the attributes given related to the complete re-inspection of supports for all welding attributes.
8608270199 860814 PDR ADOCK 05000412 G
PDR i
TE 4 'O
United States Nuclear Regulatory Commission Dr. Thomas E. Murley, Regional Administrator Systematic Assessment of Licensee Performance (SALP)
Report No. 50-412/85-98 Page 2 Each unsatisf actory condition is uniquely investigated by re-verifying the previous work accepted by that inspector.
Thus, we establish whether or not we have an isolated error.
Depend.ent upon the type of error found, we re-train the inspector, clarify our procedures / techniques, or if minor and isolated, admonish and remind the inspector of his duties.
In the specific incident mentioned in the SALP Report, (85-98) the error was completely out of character with reference to the individual's past performance as confirmed by an extensive review of his past work.
We believe that this is the first time that an inspector error was reported by the NRC during the 10 years of Duquesne Light Company Site Quality Control's existence.
The effectiveness of our actions to contain and correct potential inspector error has rece'ntly been confirmed by the independent site NRC Irspection Group, June 1986, where 42 supports and 94 hilti installations were re-inspected by random selection with no findings or errors being re->o rted. This confirmed the previous independent NRC Inspection where 151 pipe welds, 30 supports and 97 hilti installations were also re-inspected with no findir.gs or errors being reported.
We regretfully a: cept that inspector error will occur, but we believe that our training programs, supervisory actions and re-verification /sanp-ling programs are sufficient to contain this problem.
Functional Area 2 - Piping System and Supports, Function Area 5 - Electrical Power Supply and Distribution, Functional Area 6 - Instrumentation and Control System These three areas each discussed one or more inspector identified violations in which more attention to detail in either engineering or construction activities would have prevented the occurrence of the violation.
We acknowledge both the accuracy of the inspectors findings and the conclusion drawn by the SALP board.
As a result of these findings, we have taken steps to ensure that fu-ture installations. will be performed in such a manner that similar problems will not occur.
l
g
~
United States Nuclear Regulatory Commission Dr. Thomas E. Murley, Administrator Systematic Assessment of Licensee Performance (SALP)
Report No. 50-412/85-98 Page 3 We have also concluded that in some instances our field engineering personnel have had a tendency to rely upon the Hazards Analysis Walkdown group and the Environmental Qualification Survey tean to identify sit-uations in which reliable operation of equipment could be threatened by seismic events, piping failures, or high anbient temperatures, radiation, moisture and submergence to which the installed equipment might be exposed.
In addition to n2merous procedural changes, we have taken steps to improve our performance in these areas which are discussed in a sunmary at the end of this response.
Functional Area No. 8, Preoperational/ Start-up Testing A)
Concerning the QA/QC Surveillance Group which was fonned in 1985 and was being staffed by the beginning of 1986, approximately 40 people had been assigned to the Quality Assurance Surveillance Group by July 1, 1986. Additional personnel from the Operations Quality Control Department will be incorporated in the Group September 1,1986, after completion of the Beaver Valley Unit 1
~
refueling outage.
A written policy describing the division of responsibility between Site Quality Control and Test Quality Control was issued on May 16, 1986 by the Quality Assurance Manager.
Additional clarification was provided in a subsequent i
letter dated June 27, 1986, issued by-the Director of Site Qual-ity Control and the Deputy Quality Assurance Manager.
B)
The wording of Section 8, Preoperational Startup Testing implies,
that many changes were made to the Startup Program and Manuals due to NRC Audits.
However, most discrepancies were resolved by further verbal discussions / clarifications with the various NRC Inspectors. The following are examples:
J l
l r
I
United States Nuclear Regulatory Commission Dr. Thomas E..Murley, Administrator Systematic Assessment of Licensee Performance (SALP)
Report No. 50-412/85-98 Page 4 1.
Paragraph 3 states, "The Startup Manual (SUM), which is made up of individual test procedures, was not completed when first reviewed by NRC in January,1985."
i Comment:
While this statement is correct that the SUM was not totally complete when the NRC first reviewed it, the statement is incorrect in stating that the manual is made up of individual test procedures.
This was subsequently discussed with the NRC to explain that the SUM is not test procedures but the Admin-4 istrative Program / Procedure for the Startup Group (i.e., how the test procedures are written and performed, and other areas of Startup Group activities, etc.).
2.
Paragraph 3 further states, "The SUM lacked an overall program description that would tie the individual test procedures into a comprehensive format. Presently,- the individual test procedures tell the user how to execute the test program, but a program description is needed to consolidate the test program."
Commenti The intention of the SUM (not test procedures) is to direct the individual on how to execute the test program and not to provide an overall program description.
The FSAR, Section 14, Initial Test Prograra provides the overall program descrip-tion.
During subsequent meetings / discussions with the NRC Inspectors it was agreed that since the FSAR provides the over-all program description and the SUM (which references the FSAR) are the Administrative Procedures used to comply with the FSAR, no changes were needed.
1 3.
Paragraph 4 states, "The procedures were not true scheduling elements because some were so large they had to be subdivided i
and performed over an extended period of time.
Consequently significant effort was needed to form an integrated work schedule."
f Comment:
With the exception of a few test procedures (i.e. Hot.
Functional Test Procedure, Power Ascension Test, etc.) the test procedures were not intended to be scheduling documents.
- Also, j
it was further explained to the NRC Inspectors that the reason for performance over an extended period of time was the need for i
i different plant conditions (i.e., initial system performance cold and during hot plant conditions).
J
United States Nucler Regulatory Conunission Dr. Thomas E. Murley Systematic Assessment.of Licensee Performance (SALP)
Report No. 50-412/85-98 Page 5 4.
Paragraph 4 further states, " Secondly, some of the earlier pro-cedures reviewed did not have good system restoration sections or double verification of realignment of critical components."
Comment: The NRC Inspector was looking at a specific section of the ' procedure entitled " Restoration".
It was demonstrated to the Inspector that restoration was accomplished by the procedures as written by either specific steps throughout the procedure, specific steps at the end of the procedure or as directed by Operational needs through other procedures.
j 5.
Paragraph 4 also states, " Jurisdictional control over pre-test activities appeared to lack definition."
Comment: This appeared to have been an early concern of the NRC that was subsequently closed through better understanding of the Startup Group Programs.
6.
Paragraph 4 is ended by, "However, by the end of the assessment period, these concerns were receiving appropriate attention."
Consnent:
Most items of concern were resolved through dis-cussions with or better understanding of the Startup Group Programs by the NRC Inspectors. Few changes were made.
SUMMARY
The SALP suggests that "overall review and control activities are too narrowly focused on 'backend' activities" and that there should be "more licensee aggressiveness in self-identification of problems with 'frontend' activities." The NRC staff should be aware of numerous project initiatives which have been directed specifically to the early identification and resolution of problems.
Configuration Control / Advance Change Notice Systems:
Project Procedure 2BVM-56A has been developed to obtain complete configuration control through management of design change; to provide to construction, inspection, and test personnel information sufficient to schedule implementation of changes; and to provide to test and operations personnel, documents which are current and reflective of as-built, as-tested conditions.
The program utilizes an engineering checklist which requires that the engineer consider the impact of the change upon each discipline.
Thus, interface with other disciplines are identified "up-front".
1
r United States Nuclear Regulatory Commission Dr. Thomas E. Murley, Regional Administrator Systematic Assessment of Licensee Performance (SALP)
Report No. 50-412/85-98 Page 6
, Pipino Systeras and Supports:
Examples of self initiated "frontend" engineering activities which were directed towards concerns related to catalog hanger components included: i) testing of clamp anchors to evaluate the localized effect of the clamping action ii) oversized clamp holes 'were identified and a pragram was established t'o test and evaluate the condition, and iii) cracked spherical bearings were also identified as a potential problem and a program has been implemented to examine the installed condition.
Pre-Cable Pull Walkdown:
The entire routing of installed raceway for planned cable pulls is walked prior to the issuance of cable pull tickets to craftsmen. This step supplements the documentation research in identifying and correcting any deficient conditions prior to initiating the cable pull.
Raceway Installation Support Program (RISP):
The RISP personnel are located in the buildings.
They interf ace di-rectly with the craf t to resolve interferences and questions related to conduit installation.
In addition, the RISP reviews conduit drawings for possible problems.
Dry-Fit Programs:
Instrument tubing is fit up in the " dry" condition.
Thus, field interferences are identified and corrected prior to permanent installation.
Field Investigation of New Design (FIND):
Field personnel walkdown newly issued design drawings. This group was utilized for both mechanical and electrical installations.
The group identifies potential field problems with the drawings and resolves these problems.
1
r t
United States Nuclear Regulatory Commission -
Dr. Thomas E. Murley, Regional Administrator Systematic Assessment of Licensee Performance (SALP)
Report No. 50-412/85-98 Page 7 Start-Up Support Group (SUSG):
These engineers have the responsiblity of providing engineering sup-port to the DLC/SUG construction proof testing and Pre-Op/S0V testing programs.
Working with the'DLC/SUG test and system engineers, SWEC.SUSG provides engineering direction and works with the SEG in reviewing and resolving DLC/SUG test program concerns. Once the test-ing priorities are ' established, the SUSG will initiate or follow through with the SEG on all issues which require engineering support or action.
Quality Improvement Management Programs:
Under the QIMP, installation inspection results are closely reviewed to recognize the factors consistently contributing to inspection "unsats" and implementing appropriate corrective measure to improve the inspection acceptance of installations.
Since the progra.n initiation, significant improvement has been attained in the final inspection acceptance of pipe and instrumentation raceways and terminations.
Also, under this program compliance with the Rework Control Procedure is closely monitored.
Any adverse trends are addressed at the Construction Supervisor level for appropriate correction, including termination of workers found disregarding the control procedural re-quirements.
In Conclusion We are generally pleased with the SALP Board's assessment of our per-formance and have discussed with site personnelyour observations which would be of benefit to the overall excellence of project activities.
DUQUESNE LIGHT COMPANY By VJ. J: Carey '
Vice President RJW/ijr cc: Mr. P. Tam, NRC Project Manager Mr. W. Troskoski, NRC Senior Resident Inspector Mr. L. Prividy, NRC Resident Inspector NRC Document Control Desk