TXX-4180, Forwards Response to 840424 & 0501 Ltrs Containing List of Allegations Re Certain Practices.Responses Based on Knowledge of Actual Job Site Conditions.Response Includes 16 of 24 Allegations.Evaluation Continuing Re Remaining Eight
| ML20091B695 | |
| Person / Time | |
|---|---|
| Site: | Comanche Peak  |
| Issue date: | 05/25/1984 |
| From: | Clements B TEXAS UTILITIES ELECTRIC CO. (TU ELECTRIC) |
| To: | Eisenhut D Office of Nuclear Reactor Regulation |
| References | |
| TXX-4180, NUDOCS 8405300371 | |
| Download: ML20091B695 (24) | |
Text
.
TEXAS UTILITIES GENERATING COMPANY MKYWAY ToWP;H
- 400 NOHTH t bl.IVF: MTHF:P:T l..II. M t
- IBAI I.Ah, Tt;x Ah 75; tut
".;';I. ".fli"L".L%.
,,o May 25,1984 TXX-4180 Mr. Darrell G. Eisenhut, Director Division of Licensing U.S. Nuclear Regulatory Commission Dockets:
50-445 Washington, DC 20555 50-446 Comanche Peak Steam Electric Station Units 1 and 2 Allegations Transmitted by Letter of April 24, 1984
Dear Mr. Eisenhut:
This letter responds to your letters of April 24, and May 1,1984 containing a list of allegations about certain practices at Comanche Peak Steam Electric Station. To aid in an understanding of our responses, we have repeated the allegation as stated to us in your letter of April 24 followed by our response in the format requested.
Our responses were developed based on knowledge of actual job site conditions, review of existing documentation found to be analogous to certain of the allegations, and in some cases interviews with cognizant personnel.
As you will note, we have responded to 16 of the 24 allegations. We have not concluded our evaluation of the remaining eight at this time.
Our response to those will be submitted by June 1,1984.
Where applicable and as noted in the appropriate response, existing documentation supporting our evaluation is available for your review. We elected not to submit this documentation at this time to avoid burdening your review process. Relative to your letter of May 1, we have made no changes to our QA Program as a result of this particular evaluation.
However, in our assessment of these and other allegations, we have detennined that many may have esulted from a lack of proper communication between management and the workers.
In our efforts to address this TUGC0 management is continuing to promote and work toward improved conmunications at all levels at CPSES.
Ch!5 f
A PDR a unvsusou or ruun ersu.ursan nacinue cous.Asr
To the best of our management's knowledge, no documentation relating to this evaluation has been discarded or destroyed.
We trust you will find the enclosed infonnation helpful in expediting closure of these issues. Please advise if you require further information.
Very truly yours, BRC:ln cc: Mr. John Collins U. S. Nuclear Regulatory Commission Region IV i
a l
i i
TXX-4180 5/25/84 Allegation No. 2 It has been alleged that craft personnel made unauthorized design document changes in the field. They did this by writing a " traveler" which allowed them to use an incomplete design document package in the field.
Evaluation of Validity Consistent with Criterion V of 10CFR50, Appendix B, the " traveler" system is prescribed by Brown and Root Procedure CP-CPM-6.3, entitled
" Preparation, Approval, and Control of Operation Travelers." This procedure as well as the governing document control procedure DCP-3, provide for including with the traveler only those documents required to perform the work. Further, both procedures require documents used in this manner to be stamped for use only in conjunction with the operation traveler, thus precluding the document from being used in a manner other than that intended by management.
Safety Significance None Generic Implications on Other Systems or Contractors Not applicable
TXX-4180 5/25/84 Allegation No. 3 It has been alleged that the document control computer is not accurate in that it does not match the documentation.
Evaluation of Validity In any situation where information is manually coded into a computer, the possibility for input error is present.
In fact, several Nonconformance Reports (NCR's) have been issued by QC on this subject, and a Stop Work Order was issued in October, 1983. Appropriate remedial action was taken by Construction Management, and the Stop Work Order was rescinded five days later. This illustrates that the QA Program functioned properly.
Since the original identification of the issue by QC, an increased awareness of the potential for this type situation to exist has focused increased attention by Construction Management and QC on the issue.
Because of this increased attention which includes a group established to spot-check the computer data base against the field document packages, we believe the probability for these isolated errors to go l
undetected is very small.
Errors which are discovered are, of course, corrected.
Safety Significance None j
Generic Implication on Other Systems or Contractors l
Not Applicable l
l
TXX-4180 5/25/64 Allegation No. 4 It has been alleged that instances have occurred where documents were lost. No log was kept of lost documents, and on occasion craft and management would find a way around the missing documentation. An example is they would call up a missing control modification card (CMC) on the computer and delets it.
Evaluation of Validity We have assum2d that the term cocumentation refers to design documents.
Our evaluation of this allegation is that it is partially true, but without technical significance. On a project of the magnitude of CPSES with the volume of documents required for construction, some documents m';y be misplaced or lost. Although logs of lost documents have not been kept, there is a file of requests for replacement documents maintained in the Document Control Center.
We have not been able to define how craft and management could " find a way around the missing documentatic,n" short of replacing it.
The example cited lacks specificity, and, accordingly, a direct response is not possible. There are sound reasons for withdrawing information (including CMC's) from the computer data base, e.g., change in design, deleted components, etc. Without further information on this allegation, no more definitive response can be made.
Safety Significance None Generic Implications on Other Systems or Contractors Not applicable
TXX-4180 5/25/84 Allegation No. 6 It has been alleged that original, permanent records (i.e., weld data cards) are not being stored in a fire proof vault.
Evaluation of Validity Inprocess documents are not stored in a fireproof vault. These documents are stored in the field at inprocess record stations while at some stage in the cycle between initiation of the record and the final required sign off by Quality Control personnel. When they are completed, they are transmitted to the storage facility described in Chapter 17.1 of the FSAR.
Safety Significance None Generic Implications on Other Systems or Contractors Not Applicable I
1 l
t
~
TXX-4180 5/25/84 Allegation No. 7 It has been alleged that hundreds of packages of permanent records have been lost in the course of various moves.
Evaluation of Validity This allegation is partially valid but is without technical significance. The facts associated with this matter are as follows:
ASME ITEMS 1.
In Mid 1983, a task force approach to the management of ASME Component Supports was implemented. These groups were responsible for assuring that the design, fabrication, installation, inspection and documentation assembly activities were accomplished correctly and in proper sequence. The task forces were self sustaining and were organized to include a group whose only responsibility was to initiate and/or track documentation packages.
2.
Following completion of the task force activities for a given support, the completed documentation package was transmitted to the QA department for review by QA personnel and subsequently by the Authorized Nuclear Inspector.
Following satisfactory reviews, the packages were transmitted to the Construction Records Vault for storage.
3.
The ASME Boiler and Pressure Vessel '^34,Section III, Subsection NA requires the Certificate Holdcd (Et wn & Root) to s. rtify that installation aspects of the lad..3r
,een satisfied.
At CPSES, this effort involves among othtr things, verification that the documentation that has been generated and completed has been concurred with by the Authorized Nuclear Inspector and is on file in the Construction Records Vault. Appropriate controls were established as part of this verification effort to assure that unauthorized removal of documentation packages from the vault would
TXX-4180 5/25/84 not occur.
If documentation can not be located, a Nonconformance Report is issued and resclved in accordance with established procedures. Typically, such situations have resulted in the removal of the supports (or portions thereof) and replacement. Out of approximately 16,895 ASME supports for Unit 1 and Common at CPSES, on the order of 15 have been totally or partially refabricated because of lost documentation.
OTHER THAN ASME ITEMS A program similar to the verification effort described above for ASME items was also established and implemented for the Non-ASME documentation at CPSES.
Deper'ing on specific details, resolution of documentation deficiencies is achieved through reinspection and documentation and occasionally some rework. The exact magnitude of these activities is difficult to extract from the record system, but is considered minimal.
Safety Significance None Generic Implications on Other Systems or Contractors A natural outgrowth of the various activities associated with documentation verification has been centralization of record tracking activities and management emphasis on detail.
In view of the controls described above, this is obviously an economic issue which does not in any way compromise the safety of the plant.
i TXX-4180 5/25/84 Allegation No. 8 It has been stated that document control clerks are issued " controlled stamps" which they use to certify that a document package contains the latest information and is ready to be used to perform the work or the inspection.
It is alleged that these stamps were issued to the Quality Control Department, by management, and that they would stamp their own drawings and declare them legitimate.
Evaluation of Validity Our investigation of this allegation indicates that control stamps were used by Brown & Root ASME QA personnel for a short period of time.
These stamps were used to stamp a drawing to be used by the Authorized Nuclear Inspector (ANI) in reviewing hanger packages. The stamping was done in order to provide a red stamp for the ANI's, as required by l
procedure.
It should be noted that these drawings were used only for review and not for construction.
Safety Significance None l'
Generic Implications on Other Systems or Contractors Not Applicable l
l l-l l
I
TXX-4180 5/25/84 Allegation No. 9 It has been alleged that sometimes documents are outright falsified.
An example is that a date was changed on a weld data card by a quality control (QC) inspector.
Evaluation of Validity We know of no know instances in which entries on a weld data card have been falsified. There was a case in which a date on a weld data card was changed by a QC inspector. This discrepancy in dates was identified by the document reviewer, and an NCR was issued.
Investigation of this matter at the time it occurred concluded that the QC inspector was confused on dates, and there was no intent to falsify
-the record.
Safety Significance None Generic Implications on Other Systems or Contractors Not applicable i
l
TXX-4180 5/25/84 Allegation No. 10 It has been alleged that craft would " bootleg" rework by performing repairs without any documentation.
Evaluation of Validity We know of two situations where this has happened. Darlene Stiner testified before the ASLB that she repair-welded small bore pipe supports without proper documentation, and our investigation confirmed that she did so.
However, these repair activities were detected by QC, and the hangers were scrapped. These facts were presented in the hearings before the ASLB.
Further, NRC Region IV has identified three cable tray supports in the Unit 2 Cable Spread Room out of 87 inspected, which are repair welded and for which we are as yet unable to locate repair documentation.
We are continuing to explore this matter and will report the results and implications to the Staff shortly in response to Mr. Bangart's letter of April 30, 1984.
Our investigation revealed no other instances of undocumented repairs which were not identified and dispositioned per applicable procedures.
Safety Implications None Known Generic Implications on Other Systems or Contractors Not applicable
TXX-4180 5/25/84 Allegation No. 11 Describe the utility and' contractor programs for verifying that the "ac-built" conditions accurately reflect design.
It has been alleged that ir.stt cas hcVe occurred where craft, instead of following the appropriste document (i.e., drawing, NCR, CMC), would make things fit a
and have an engineer write "as-built" on the document. Then the actual "as-built" condition would not be sent 'ack to Gibbs & Hill for evaluation. An example is a nonconfSemance report (NCR) that was written because material did not meet minimum wall thickness and an engineer voided it by writing "as-built".
Evaluation of Validity Our evaluation indicates that the allegation is in part correct.
Instances have occurred where the craft, unable to construct to the approved design due to interferences in the field, has "made things i
- fit" and sought engineer approval of the "as-built" condition on a
-Design Change Authorization (DCA) or a Component Modification Card (CMC).
In such situations engineering conducts its independent review of the "as built" condition, and may accept it or reject it.
In cases where the "as built" condition is considered adequate, design change u
[
documents (DCA's and CMC's) are returned to the original designer (normally Gibbs & Hill) for design review. To further assure that the as-built condition conforms to the latest design, Quality Assurance has performed a design change verification in which they confirmed that inspection records are available which reflect that Quality Control has inspected installations in accordance with the latest design, including applicable design changes.
Our complete review of voided Nonconformance Reports (NCR's) for both Brown & Root and Texas Utilities indicates that in no case has an NCR been voided by an engineer. Voiding of an NCR by an engineer is prohibited by both TUGC0 and Brown & Root QA programs. Both programs require that if an NCR is to be voided, QA/QC concurrence must be obtained and justification for voiding be provided on the NCR.
TXX-4180 5/25/84 Safety Significance None Generic Implications on Other Systems or' Contractors Not applicable I
d Y
I l
l-L
[:
i i
i t
--v.
TXX-4180 5/25/84 Allegation No. 12 It has been alleged that Quality Control inspectors are not qualified or have insufficient training.
Evaluation of Validity Our evaluation of this allegation indicates that the training and certification programs (ASME and Non-ASME) at Comanche Peak are in full compliance with ANSI N45.2.6, SNT-TC-1A, and Regulatory Guide 1.58.
These programs require an evaluation of an inspector's education, experience, and formal training. Quality Control supervisory personnel evaluate an inspector's proficiency prior to recommending certification. This evaluation of ability continues after certification.
If any situation develops which causes supervision to question an inspector's ability to perform an inspection activity adequately, that person is removed from that activity and work inspected by him is reinspected.
In isolated instances, individuals have performed inspections for which they were not properly certified.
Those instances have been documented on nonconformance reports, and appropriate corrective action has been taken.
Periodic surveillances and audits conducted by TUGC0 and Brown & Root ensure compliance.
Various outside agencies such as INPO, CAT, NRC Region IV, Hartford Steam Boiler, and ASME have evaluated and concurred with the training and certification of inspection personnel.
Safety Significance None Generic Implications on Other Systems or Contractors Not applicable
TXX-4180 5/25/84 Allegation No. 13_
What is the purpose of the N-5 program? Could the program hinder the document controller from carrying out his/her job?
It has been alleged that the N-5 program is making it difficult for document controllers to review records in the records vault when necessary and makes it impossible to review packages going into the vault.
Evaluation of Validity The purpose of the N-5 program is to describe Brown & Root's method of compliance with requirements established in ASME III, Division 1, Subsection NA-3460.
In doing so, Brown & Root has established a QA program which has been approved by both the ASME and an Authorized Inspection Agency (Hartford Steam Boiler Inspection and Insurance Co.).
This program describes methods of conducting and documenting piping system installation in accordance with the Code. The penultimate step in this program is a Brown & Root QA review of records documenting the piping system installation, followed by certification of the N-5 Code Data Report certifying Brown & Root QA acceptance of the installation and compliance with the ASME Code.
The term " document controller" as used in the allegation may be misleading. We believe that the job description referred to in the allegation is one of a " document reviewer" who is a person trained by Brown & Root QA in methodology in conducting document review described above.
Performance of this review requires interface between the i
document reviewer and the Permanent Plant Records Vault (PPRV). At times this interface may be delayed by the computer indexing and coding of records for the Automated Records Management System (ARMS). Upon receipt of records into the PPRV, TUGC0's QA procedures require that documents be forwarded to the ARMS Coders to provide an indexing system for the record which accomodates record retrieval.
If a specific record requested by a dccument reviewer was in the coding process, it would not be released to the reviewer until coding was complete. We
4 TXX-4180 5/25/84 can see how this could be frustrating, but fail to see how this is
" hindering" the document reviewer. The internal PPRV record processing could delay the document reviewer's efforts for a short time.
- However, as the document reviewer's function is to review records prior to N-5 certification, and since this review is programmatically mandated prior to N-5 certification, the allegation appears without technical or regulatory merit.
Safety Significance None Generic Implications on Other Systems or Contractors Not applicable I
I l
l
TXX-4180 5/25/84 Allegation No. 15 Describe the utility and contractor programs for receiving vendor parts and certification of vendor components.
It has been alleged that if parts were lost from a vendor component, the fabrication shop manufactured replacements without procedures. The practice of fabricating "Q" (safety related) material without quality control oversight occurred regularly.
' Evaluation of Validity Vendor parts are received in the same manner that all material is received at CPSES. Material is received by the Material Control personnel in accordance with Procedure CP-CPM-8.1.
For safety related items, Material Control personnel then notify Receiving QC to perform receipt inspection.
This is performed in accordance with Quality Instructions (procedure number dependent upon type of material or component being received). Part of this receiving inspection is to ensure:
- 1) That required vendor documentation accompanies the material, part, or component;
- 2) That the vendor document matches the material; and
- 3) The vendor documentation is in accordance with applicable codes, standards, specifications, and purchase order requirements.
Our investigation of this allegation indicates the fabrication processes utilitized to fabricate some parts, i.e., bending, threading, cutting, welding, are addressed by site construction procedures, and required inspections are mandated by quality control procedures / instructions. For specialty type vendor supplied items which were refabricated on site, specific instructions were provided by Engineering on a Construction Operation Traveler, and vendor concurrence was obtained.
.a
=
a
.2
...-4._2
_a-..
&-_a__a
.* ~.
g TXX-4180 5/25/84 In sumary, fabrication practices are properly controlled by construction and inspection procedures and instructions. Our investigation did not uncover instances where the fabrication shop manufactured safety related replacements without procedures or without QC oversight.
Safety Significance None Generic Implications on Other Systems or Contractors Not applicable I
t i
i 4
F i
i 3
TXX-4180 5/25/84 Allegation No. 21 It has been alleged that safety related welds were repaired with weld tech (W.T.) holdpoints, instead of QC hold points in violation of procedures. What are the effects of this?
Evaluation of Validity The facts in this allegation are correct; however, this practice is not prohibited by procedure.
In fact, procedures endorse this practice.
Prior to initiating a weld repair, Welding Engineering prepares a Repair Process Sheet (RPS) describing the method of repair, including the weld procedure to be used and required welding technician and quality control hold points for inspection. QC hold points are listed on the RPS from an inspection matrix provided in a procedure approved by the Brown & Root QA Manager. Once the QC and Weld Tech hold points are specified, the RPS is sent to Engineering where required, to Westinghouse for review (for Westinghouse components only), and to the ANI for review and assignment of ANI hold points, all of which occur prior to issuance to the craft. Weld Tech hold points are those added at the discretion of the Welding Engineering department, but are not in lieu of QC required hold points. Consequently, no procedure violation exists.
Both Brown & Root and TUGC0 NCR logs were searched for this type of violation. The only related instance (although not substantiating the allegation) was detected by QC, documented on an NCR, and corrective l
action taken.
In this case, a welding technician signed what was indicated to be a QC hold point.
In fact it had been incorrectly designated.
It should have been a weld tech hold point.
Safety Significance None Generic Impact on Other Systems or Contractors Not applicable
TXX-4180 5/25/84 Allegation No. 22 Although the Quality Control (QC) and Construction organizations are represented as independent, it has been alleged that current practices between QC and construction compromise the independence of these two organizations.
Evaluation of Validity Our review of this matter indicates that there is no validity to this allegation. Although the Building Management Groups are " matrix" organizations, the Building QC organizations report to QA/QC management and not construction.
Each Building QC Supervisor was selected based on demonstrated ability to work within the Matrix organization and still maintain the independence required for QC personnel.
We have found no evidence which lead us to believe that the alleged lack of QC independence exists.
Safety Significance None Generic Implications on Other Systems or Contractors Not applicable
TXX-4180 5/25/84 Alleoation No. 23 It has been alleged that craft bypassed procedures by telephoning orders to the fabrication shop in lieu of sending drawings.
Evaluation of Validity Our investigation of this allegation indicates that the allegation is partially correct, but not significant. Occasionally, requests for shop fabrication are relayed to the shop foreman by telephone. This is not in violation of site procedures.
There is no need for drawings to accompany the request for fabrication, as the fab shop has access to any drawings that may be required.
Safety Significance None Impact of Other Systems or Contractors Not applicable
r-TXX-4180 5/25/84 Allegation No. 24 It has been alleged that there is constant pressure, by craft and management, on QA/QC inspectors not to write non-conformance reports.
Evaluation of Validity This allegation is not valid. The procedures governing initiation and processing of Nonconformance Reports have been established to preclude anyone (including management or supervisors) from stopping an inspector from initiating a Nonconformance Report if in his judgment, such a l
report is warranted. The inspector has only to call a clerk to obtain an NCR number, and the report must then be entered into the system and
)
resolved.
1 CPSES also uses other methods to document conditions not in compliance with specified requirements, specifically Inspection Reports or in the case of electrical separation or some final inspections, Deficiency Reports (listing of work items required to be completed to achieve conformance with specified requirements). The basic difference between i
the use of these documents and a Nonconformance Report is that these l
latter documents do not routinely involve the use of " Hold Tags." If l
)
items identified in this manner cannot be made to conform to requirements, a Nonconformance Report will be issued and the matter resolved per established procedures.
r Additionally, there have been a series of 01 investigations within the i
past year relative to this same subject and none of these investigations has indicatea' a practice similar to the allegation.
There has never been any pressure on QA/QC inspectors not to report nonconforming conditions in the manner and on the proper form prescribed by procedures.
Nonconforming conditions may be reported and l
documented on Nonconformance Reports, Inspection Reports, and l
Deficiency Reports. There have been instances when QC inspectors have l
P
TXX-4180 5/25/84 been reminded by QC management of the need to comply fully with procedures, including the need for inspectors to report nonconforming conditions on the appropriate form. There have been instances when inspectors have been instructed to report nonconforming conditions on Inspection Reports, and not on Nonconformance Reports, because the procedure provided for the use of Inspection Reports. These instances were motivated by management's desire to assure that reporting requirements are met, and certainly not by any desire to have nonconforming conditions go unreported.
Relative to alleged craf t pressure, QC Inspectors are totally independent of the construction (craft) organization. Thus, " pressure" cannot be provided.
It is possible however, to have exchanges of feelings from time to time between craf t and QC, and all levels of TUGC0 and Brown and Root management have demonstrated that such exchanges are controlled. The best example of this control can be seen through a review of Section VII of the CAT report for CPSES.
Safety Significance None Generic Implications of Other Systems or Contractors Not applicable
STATE OF TEXAS
)
COUNTY OF DALLAS Billy R. Clements being duly sworn deposes and says:
That he is Vice President, Nuclear Operations, Texas Utilities Generating Company and knows the contents of the foregoing Applicants' response to Darrell G. Eisenhut's April 24, 1984 letter transmitting allegations; that the same is true of his own knowledge except as to matters therein stated on information and belief, and as to that he believes them to be true.
n Billy R.R iements Subscribed and sworn to before me this 25 day of May 1984.
J>
49 U sanoA BEso'N. istary ruti;e in and for Dallas Co Myconunissione gty. xasg:
i
..