ML20212K826
| ML20212K826 | |
| Person / Time | |
|---|---|
| Site: | Browns Ferry  |
| Issue date: | 03/02/1987 |
| From: | Domer J TENNESSEE VALLEY AUTHORITY |
| To: | Grace J NRC OFFICE OF INFORMATION RESOURCES MANAGEMENT (IRM), Office of Nuclear Reactor Regulation |
| References | |
| NUDOCS 8703090426 | |
| Download: ML20212K826 (10) | |
Text
O TENNESSEE VALLEY AUTHORITY CHATTANOOGA. TENNESSEE 37401 SN 157B Lookout Place MAR 021987 U.S. Nuclear Regulatory Commission ATTN: Document Control Desk
~
Office of Nuclear Reactor Regulation Washington, D.C.
20555 Attention:
Dr. J. Nelson Grace In the Matter of
)
Docket Nos. 50-259 Tennessee Valley Authority
)
50-260 50-296 BROWNS FERRY NUCLEAR PLANT UNITS 1, 2. AND 3 - NRC-OIE REGION II INSPECTION REPORT NOS. 50-259/86-25, 50-260/86-25, AND 50-296/86-25 SUPPLEMENTAL RESPONSE TO VIOLATIONS 2.A AND 2.C Enclosed is our response to your January 26, 1987 letter to S. A. White requesting a supplemental response for items contained in Inspection Report
'Nos. 50-259/86-25, 50-260/86-25, and 50-296/86-25 relative to the mounting configuration of the Control Room Emergency Ventillation System skids and the welded blank found in the Preaction Fire Protection System.
For the purpose of clarification, the supplemental responses contain the information previously submitted in TVA's original response'with the additionally requested justifications and clarifications where required.
If you have any questions, please get in touch with M. J. May at (205) 729-3566.
To the best of my knowledge, I declare the statements contained herein are complete and true.
Very truly yours, TENNESSFE VALLEY AUTHORITY
/.
$P&T.
J. A. Domer, Assistant Director Nuclear Safety and Licensing Enclosures cc: See page 2 8703090426 870302 l
PDR ADOCK 05000259 o1 O
t An Equal Opportunity Employer
~
..U.S. Nuclear Regulatory Commission 0 2 1987 cc (Enclosure):
Mr. James M. Taylor, Director U.S. Nuclear Regulatory Comission Office of Inspection and Enforcement Washington, D.C.
20555 Mr. G. G. Zech, Director U.S. Nuclear Regulatory Comission TVA Projects 101 Marietta Street, NW, Suite 2900 Atlanta, Georgia 30323 Browns Ferry Resident Inspector Browns Ferry Nuclear Plant Route 2, P.O. Box 311 Athens, Alabama 35611
SUPPLEMENTAL RESPONSE NRC INSPECTION REPORT NOS.
50-259/86-25, 50-260/86-25, AND 50-296/86-25 J. N. GRACE'S LETTER TO S. A. WHITE' DATED JANUARY 26, 1987 BROWMS FERRY NUCLEAR PLANT (BFN)
ITEN 2.s ORIGINAL VIOLATION 10 CFR 50, Appendix B, Criterion V, requires that activities affecting quality shall be prescribed by documented instructions, procedures, or drawings of a type appropriate to the circumstances and shall be accomplished in accordance with those instructions, procedures, or drawings.
Contrary to the above, this requirement was not metrin that the fire.
protection proaction sprinkler system on the 593 foot elevation of the unit 3 reactor building was neither prescribed as in Plant Drawing 47W491-30 nor installed with an adequate system flush as required by note two of Drawing 47W491-1. Note two of the drawing states that fire protection pipes
-shall be cleaned free of scale, dirt rust, and welding icicles, flushed with treated water and blown out with compressed air prior to final closure.
Failure to remove the welded plug and perform an adequate flush as required resulted in the sprinkler heads being unable to provide water spray to the cable trays and equipment in the area. This condition was thought to have existed since 1977.
REQUEST FOR SUPPLEMENTAL RESPONSE 1
Clarification and a supplemental response is required to your reply to Violation 2.a.
For corrective action, you have stated that the Fire Protection Group is initiating a procedure to perform special air testing at selective locations on all critical structures, systems, and components preaction sprinkler systems to verify open flow paths.
Further, this test will verify flow through selected header sections and their associated branch lines. However, the Licensee Event Report 50-296/86006 covering information j
on the same subject states that a 30 percent random sample of branch lines of the proaction sprinkler systems was verified for open flow path. The LER provides a quantitative response and it is not clear to us if it is consistent j
with the former information. Please provide clarification. In addition, provide justification for not performing a 100 percent inspection of the preactive sprinklers. If a certain sample size was indeed selected for inspection, then provide the basis for its selection and why it was considered adequate to insure no flow blockage.
1 J
l i
. Item 2.a (SUPPLEMENTAL RESPONSE) l~
2.
Reasons for the Violation
.In December 1976, multiple workplans were issued to install the remaining proaction sprinkler system in the unit 3 Reactor Building.
Due to inadequate coordination in workplan preparation, a blank installed to flow test the main system header addition was not removed when the branch line piping for the northwest corner of the Reactor Building at Elevation 593 was installed and hydrotested.
(The blank was utilized in two separate workplans, neither workplan required its removal.) It was mistakenly believed the flow test on the main system, in addition to the hydrotest on the branch lines, was an adequate test for the installed piping. This was the reason the blank was not found during the 1977 postmodification test. The modification was completed, and the system declared operable on December 8, 1977.
+
Technical specifications do not require an inspection of the preaction system dry sections which is the reason the blank went undetected.
National Fire. Protection Association recommended practices on sprinkler system maintenance suggest sprinkler system flushing at 10-year intervals (NFPA 13A 1981 edition, paragraph 5-4.2).
TVA would have implemented this practice in 1987. However, a design modification scheduled for the unit 2 preaction sprinkler system will require a hydrostatic test (hydro) of the entire preaction sprinkler system as a postmodification test. To detect possible problems before the hydro, a maintenance request was initiated to pneumatically test the existing preaction sprinkler system dry P pe in the unit 2 Reactor Building. As a result of continuing this i
effort into units 1 and 3, the installation of the blank in the unit 3 preaction system was identified on July 14, 1986.
f 3.
Corrective Steps Which Have Been Taken and Results Achieved j
The preaction fire protection sprinkler systems are normally dry. The l
section of dry pipe containing the blank was removed and replaced with a new section of pipe. All preaction system dry cross mains have been tested to verify the existence of an open flow path.
The control and documentation of work performed has been improved through procedure revisions since the 1977 modifications were performed. Also, a more detailed initial review of workplans is now performed before work authorization.
In addition, all required temporary alterations occurring to existing system configurations during workplan performance are controlled by plant procedures.
In the removal of such temporary alterations, second party verification is required.
. ~...
l i A program was recently implemented which requires that all modifications and postmodification testing pertaining to fire protection systems be reviewed by the Fire Protection Unit. The Fire Protection Unit is directly involved in the postmaintenance testing of modifications associated with fire protection piping. All changes made to any procedure that pertains to a fire protection system are reviewed by the Fire Protection Unit.
Ah approximate 30-percent random sample of preaction system branch lines were selected for a special air test. This test was performed in accordance with written procedures developed by the Fire Protection Unit and was intended to identify blockages in the system's branch lines. Of the more than 100 branch lines tested, no blockages were identified. The testing was completed on November 30, 1986.
4.
Corrective Steps Which Will Be Taken to Avoid Further Violations TVA considers that the programmatic controls currently in place would not allow the reoccurrence of the 1977 event. Because of the current program, no further corrective steps are anticipated.
5.
Date When Full Compliance Will Be Achieved The justification for sample size and requested clarification is presented in the following paragraphs. is a simple sketch of one hypothetical preaction fire protection sprinkler system which is included to provide uniformity in interpreting the nomenclature used in TVA's response. A brief description of the system is further presented. The preaction fire protection sprinkler systems are normally dry. A preaction valve isolates the dry system from the charged fire main.
The valve automatically opens and charges the cross main and branch lines upon receipt of an actuation signal, smoke, or fire. Manual activation resulting from direct operator action will also open the valve. A seal on each sprinkler head must melt before it can complete its function.
Preaction Fire Protection Systems provide required fire protection for various critical structures, systems, and components. The branch lines are all screwed pipe connections from the cross taain to the sprinkler heads.
The original selection of a 30-percent random sample of preaction branch lines was made by the Fire Protection Unit based on engineering judgement.
The decision was supported by the fact that the preaction system branch lines are screwed pipe and, therefore, the possibility of finding a welded blank or plug is not feasible. As a result of NRC's request for further information on the choice of sample size and to provide justification for not performing a 100 percent inspection of the preaction sprinklers, TVA provides the following information.
_4 a.
The Probabilistic Risk Assessment Organization of TVA was asked to statistically determine what the probability of not finding a blockage in a branch line not tested would be.
Based on TVA's sample size and using standard statistical techniques, TVA has a 99-percent confidence level that 95 percent or greater of the lines not tested are-free of blockages, b.
Where possible, the special air test performed utilized a drain valve off the cross main, as the supply point for the air. Randomly selected sprinkler heads were then removed and air flow was verified at those points. Again, no blockages were identified, c.
The branch lines are all screwed pipe and the original installation i
required the piping to be cleaned free of scale, dirt, and rust and j
flushed with treated water and blown out with compressed air before final closure of the piping system.
In summary, TVA considers the error in the workplans which resulted in the blank remaining in the system to be an isolated case of improper control placed on those specific workplans. No evider.ce to the contrary has been identified through field testing of the Preaction Sprinkler System. TVA, therefore, considers this item closed.
ITEN 2.c ORIGINAL VIOLATION i
l 10 CFR 50, Appendix B, Criterion V, requires that activities affecting quality shall be prescribed by documented instructions, procedures, or drawings of a type appropriate to the circumstances and shall be accomplished in accordance with those instructions, procedures, or drawings.
i Contrary to the above, there is no documented instruction, procedure, or drawing which prescribes the anchoring details of the control room emergency ventilation system (CREVS) filter train and blower assembly. The train A and B assemblies are mounted differently with train B cantilevered off its anchoring pad.
Since no anchoring details could be located, the installed configuration represents an unanalyzed condition.
This is a Severity Level IV Violation (Supplement I) and is applicable to all units.
i l
i l
]
l l
... = -
. Request Fcr Supplemental Response After careful consideration of the basis for your denial of Violation 2.c, we agree with your position from the view point of Quality Assurance records retrieval problem only. The subject equipment was installed prior to 10 CFR Part 50, Appendix B becoming effective and consequently, some of the component installation details are not retrievable. However, you have not.
i adequately demonstrated in your response that the actual as-built equipment configuration is seismically qualified.
TVA was unable to retrieve the QA records for CREV installation. TVA's reason for this is that 10 CFk Part 50, Appendix B came into being after some of the i
j components of Browns Ferry facility were already installed, and TVA took exception to Appendix B criteria IV, XVII, and IVIII as stated in a response to AEC question D.3, detailed in the FSAR, Amendment 31.
While we agree with TVA's position on the QA documentation issue, we do not have the assurance nor accept TVA's current response in demonstrating the CREV unit's ability to maintain integrity during and after a seismic event.
In the early AEC questioning period, TVA's response to an AEC question was as follows:
i l-
"The systems have been designed, fabricated and erected to the same codes, standards, criteria and practices that would have otherwise been required l
except that QA documentation is not available."
l Furthermore, TVA has stated in the FSAR (Chapter 10.12) that the CREV units l
(Emergency Pressurization System per the FSAR) are seismically qualified.
I With regard to the seismic question, the licensee used their Division of i
Nuclear Engineering (DNE) for support in determining the CREV units' seismic qualification.
The NRC obtained a copy of the DNE evaluation and learned that the evaluation process is generic and intended to apply to all electrical equipment. The evaluation methodology is -judged to be unacceptable because it i
had been shown to be faulty over the last several years.
For example, the equipment evaluated by TVA in 1981-1982, and judged to be acceptable, has been found to be unacceptable in 1985 and 1986. This includes battery racks, control room panels, RPS instrument racks, and hydraulic control unit. These l
findings compromised the validity of conclusions drawn from the TVA activities associated with IE Notice (IEN) 80-21, Anchorage and Support of Safety-Related Electrical Equipment.
Furthermore, DNEs use of the Seismic Qualification i
l Utilities Group (SQUG) report related to the NRCs resolution of Unresolved Safety Issue A-46 concerning seismic qualification of equipment is not fully appropriate to this case. The referenced SQUG/A-46 report is too general and it provides no information or specific guidelines on acceptable anchoring of compenents such as the CREV units, even on another component similarity basis.
i i
I
. ITEM 2.c (SUPPLEMENTAL RESPONSE)
In response to the Atomic Energy Commission's FSAR questions dated l
March 25, 1971 TVA addressed the conformance of 10 CFR 50, Appendix B, in relation to specific BFN systems. The response to question D.3.
detailed in the FSAR, Amendment 31,-stated exception to criteria IV, XVII, and XVIII as they relate to the control room air conditioning system, of which CREVS is a part. This is due to the fact that Appendix B came into being after some components of BFN were already installed.
Since an exception was taken to 10 CFR 50, Appendix B, Criteria XVII, Quality Assurance Records for fabrication, installation, and examination, complete quality control or quality assurance documentation was not required to be maintained for this equipment, and thus, the anchoring details could not be retrieved in this case.
The Division of Nuclear Engineering (DNE) has initiated a problem report documenting the fact that the civil drawings do not show a design for the pads or connections for the vendor equipment. This problem report is presently being resolved.
TVA's BFN is identified in NUREG-1030 as one of the operating plants to be reviewed to the requirements of unresolved Safety Issue (USI) A-46.
The CREVS along with other components are contained within the scope of items that will be reviewed against the guidelines of USI A-46.
The SQUG guidelines to inspect and evaluate equipment anchorages are 1
currently being reviewed by the Senior Seismic Review Advisory Panel (SSRAP).
It is anticipated that the guidelines will be approved by NRC in the fourth quarter of 1987.
When the guidelines are approved. SQUG will train utility personnel as to how the guidelines are to be implemented, and upon completion of the training, TVA will develop and submit to NRC a schedule for the implementation of these guidelines.
The Engineering Mechanics Group (EMG), DNE, is responsible for determining the seismic acceptability of components and equipment at BFN and has received specific training relative to TVA's position on long-term seismic qualification of safety-related equipment.
An ENG representative has performed a field inspection of the CREV units (both system A and system B).
Additionally, on this inspection the ENG representative was accompanied by a BFN DNE Supervisor-Civil who is actively engaged with the SQUG organization in the development of the j
guidelines for evaluating anchorages to resolve USI A-46.
The ENG representative's conclusions follow:
1 Af ter consideration of the as-installed configuration, no reason could be identified to take exception to the A-46 qualification i
position of seismic acceptability in the absence of evidence to the contrary.
I
The BFN DNE Supervisor-Civil is also of the opinion that these particular installations will be qualified when the guidelines are finalized after the SSRAP review.
It is TVA's position supported by these field inspections, that because the CREV units were installed relative to existing practices for seismic installations at the time of initial construction, that there is no reason to treat the seismic reverification of these units differently from those of any other BFN system.
0630c i
l l
l l
g s
ATTACHt'ENT 1:
SIMPLE PICTORIAL REPRESENTATION OF ONE PREACTION FIRE PROTECTION SYSTEM O
i TtlREADED -
CONNECTION g
-- r O - ~ ~ ~~~~
~
O SPRINKLER t!EA1>S (TYPICAL) g-y BRANCli
{,-j_
_ g
- Cf gl_
LINES i
i j
j j_
[
i g
CROSS MAIN (DRY SECTIOH)
PREACTION m
3 VALVE 6
l l
l r
MAlii llEADER d