ML20205P586
| ML20205P586 | |
| Person / Time | |
|---|---|
| Site: | Browns Ferry  |
| Issue date: | 11/03/1988 |
| From: | Gridley R TENNESSEE VALLEY AUTHORITY |
| To: | NRC OFFICE OF ADMINISTRATION & RESOURCES MANAGEMENT (ARM) |
| References | |
| NUDOCS 8811080227 | |
| Download: ML20205P586 (23) | |
Text
, _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ - _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _.
TENNESSEE VALLEY AUTHCRITY CH ATTANOOGA. TENNESSEE 37401 5N 1578 Lookout Place NOV 03 N88 U.S. Nuclear Regulatory Commission ATTN: Document Control Desk Hashington, D.C.
20555 Gentlemen:
In the Matter of
)
Docket Nos. 50-259 Tennessee Valley Authority
)
50-260 50-296 BROWNS FERRY NUCLEAR PLANT (BFN) - RESPONSE TO NRC INSPECTION REPORT NOS.
50-259/88-07, 50-260/88-07, AND 50-296/88-07 This letter transmits TVA's response to NRC Inspection Report 88-07.
The inspection was on the calculation aspects of the BFN Design Baseline and Verification Program (OBVP) and was conducted at TVA's Engineering Offices in Knoxville during April 1988.
The inspection report provided closure of most open items from the previous inspection in October 1987; however, it also identified weaknesses relating to the implementation aspects of the OBVP calculation effort. contains our response to the NRC identifled weaknesses.
Please hote that TVA has taken exception to the statement that we have agreed to test the containment purge valve to open against 30 psig pressure before restart (Section 4.2.2.2.1 Concern (3)). Our response to this issue has been coordinated with P. Hearn of your staff and is included on page 7 of enclosure 1.
A summary If st of commitments is provided in enclosure 2.
The inspection report which was issued on September 8, 1988, requested that TVA provide a response within 30 days.
However, TVA did not retnive this report until September 26, 1988. An extension of the response due date to November 4, 1988, was agreed to by D. Moran in a telephone call.
Please refer any questions concerning this submittal to Patrick Carter, BFN Site Licensing, (205) 729-3570.
Very truly yours, TENN SSE VALLEY AUTHORITY a
.G diey, Ma ger Nuclear Licensing and Regulatory Affairs Enclosures cc:
See page 2 8811080227 881103 PDR ADOCK 05000259 g\\
g PNU An Equal opportunity Employer
- U.S. Nuclear Regulatory Commission M 03 28 cc (Enclosures):
Ms. S. C. Black, Assistant Director for Projects TVA Projects Olvision U.S. Nuclear Regulatory Commission One White Flint, North 11555 Rockville Pike Rockville, Maryland 20852 Mr. F. R. McCoy, Assistant Director for Inspection Programs TVA Projects Division U.S. Nuclear Regulatory Commission Region II 101 Marietta Street, NW, Suite 2900 Atlanta, Georgia 30323 Browns Ferry Resident Inspector Browns Ferry Nuclear Plant Route 12, Box 637 Athens, Alabama 35611 i.
l t
w Page 1 of 20 CIVIL STRUCTURAL AREA I
Concern 1.
Attachment F to Design Criteria Document 8FN-50-C-7100 provides only the criteria for the lower drywell access platforms, and it is not clear which attachment provides the criteria for the upper drywell access platforms.
TVA clarification is required.
TVA Response The upper drywell access platform criteria is attachment G of
[
BFN-50-C-7100 RI.
This is the general criteria for miscellaneous steel for Class I and Class II structures at 8FN.
The last sentence in section 1.1 of Attachment F to BFN-50-C-7100 specifies that, "For remainder of drywell platforms, see BFN-50-C-7100, Attachment G."
Concern i
2.
For the design of drywell access platforms, the jet terce was explicitly included in Section 12.2.2.7.2 of the FSAR as a concentrcted load but is now excluded from Revision 1 of Design Criteria Document BFN-50-C-7100.
TVA should determine if this exclusion violated the project licensing commitment.
Additional details related to this concern from Section 4.1.2.2 of report TVA must also evaluate the effect of jet loads on the electr :al cable and instrumentation to ensure that only one train would be affected by the postulated jet loads.
{
TVA Response 4
1 FSAR Section 12.2.2.7.1 Identifies loading conditions which were applied to l
the deywell platforms.
The term "jet" refers to the reaction force of l
i mitigating devices (which could be attached to the platform!.) subjected to j
pipe break loadings. Attachment F to Design Criteria BFN-50-C-7100
)
established the criteria for the lower drywell access platforms consist <at l
with the above. 'The equivalent static load (Yr) on the structure is generated by the pipe whip reaction from pipe rupture restraints attat:hed to the drywell l
steel.
l The main steam and feedwater pip 6 whip restraints at the drjwell penetrations at 180 degrees azimuth are designed to transfer rupture loajs from the process l
piping to the reactor pedestal and to the concrete at elevation 549.92 without significantly loading the drywell steel.
i I'
l
~
~.
Page 2 of 20 Primary emphasis for jet impingement protection inside the drywell was directed toward protecting primary contairment.
In addition to the recirculation, main steam and reactor feedwater system restraints, further consideration to containment protection was provided by installation of honeycomb panels on the inside surface of the drywell shell and jet deflectors over the ma h vent openings to the wetwell.
Protection of other equipment in the drywell is inherent in the plant arrangement of equipment.
Redundant systems and devices are located on opposite sides of the drywell to minimize the concerns of dynamic forces associated with a pipe break.
In support of this position, the following was submitted to the AEC in response to their questions of March 25, 1971, on the effects of pipe runture.
Response to question 4.1.4
"... special care is also taken in component arrangements to see that equipment associated with engineered safety sy:tems such as the core spray and the LPCI are segregated in such a manner that the failure of one ca^not cause the failure of the other." Additionally, "The redundant channels a reactor level and pressure sensing lines are located in the cylindrical section of the drywell 180* apart for maximum physical separation."
Response to question 5.16 "The core standby cooling systems are physically separated, both inside and outside the containment, minimizing the probability of simultaneous damage to more than one system from a missile source." Where in this case the missile source being jet impingement.
Concern 3.
TVA does not review the technical content of contractor generated civil / structural calculations.
TVA should review a sample of contractor generated civil /structeral calculations to ensure that the calculations are adequate and correct.
TVA Response TVA has taken steps to ensure an adequate level of review of contractor generated calculations.
The civil group has performed reviews of contractor design output to assure adequacy and compliance to general civil design criteria.
~.
Page 3 of 20 Concee 4.
Two pipe stress calculations (NI-3677T and NI-270lR) deviate from the requirement stated in the applicable design criteria and F3AR commitments.
TVA should revise those calculations to meet the applicable design criteria and FSAR commitments.
Additional details related to this concern from Section 4.1.2.2 of report.
Pipe stress problem N1-367-7T The NRC team found that two separcte values for the soll shear wave velocity are used in the calculation.
In the calculation for the soll strain, a value of the maximum ground velocity of 48 inches per second is used, which is not referenced, and which may not accord with licensing commitments.
Pipe stress problem NI-270-IR TVA, therefore, should access the seismic qualification documents for valves 2-FCV-70-313 and -47 to confirm that the valves are rigid, as modeled in the piping analysis, or revise the calculation in accordance with the requirement of Design Criteria Document BFN-50-C-7103 ff the valves show a fundamental frequency of less than 20 Hz.
TVA Response Pipe stress problem N1-367-7T The calculation for the EECH piping in the RHR-EECW tunnels was revised to incorporate the appropriate soll properties and site conditions. A soll shear wave velocity (Vs) of 1000 ft/sec was used in the analysis which is ;onsistent with Section C.2.1 of Appendix C to the FSAR.
This velocity was considered more appropriate for the firm clay soll condit'ons around the RHR-EECW tunnels.
The minimum Vs (250 ft/sec) represents an anomaly for the site.
This anomaly is very sof t soll that was only encountered in the area of the intake channel and subsequently excavated.
Therefore, this minimum Vs was not considered in establishing a reasonable average Vs for analysis of the piping. A normalized site specific value of 17 in/sec (OBE) was used for the peak ground velocity.
Pipe f, tress problem N1-270-IR The valves 2-FCV-70-313 and -47 were verified to have natural frequency values greater than 20 Hertz.
The calculation for this stress problem was then revised to note this and properly document a basi: for modeling these valves as being rigid.
i Page 4 of 20 i
Concern 5.
Section 4.3 of the essential calculation cover sheet, RIMS 830 880329 001, 4
i stdtes that calculations for major building structures will not be reviewed for technical adequacy until after restart.
TVA must review those calculations before plant startup.
TVA Response l
A memorandum was issued by the lead civil engineer to clarify the intent of Section 4.3 to indicate that the calculation effort will review the Phase II calculation types before restart.
New issues and revisions have incorporated this information.
Concern 6.
The NRC team reviewed two heating, ventilation, and air-conditioning (HVAC) ductwork support calculations but was not able to verify that the proper seismic load was used in the analysis nor to determine if prying action had been considered in the analysis of the bolts.
TVA should verify that the calculations include proper seismic loads and that prying action is considered in the analysis of the bolts.
l TVA Response i
j The HVAC Calculation, CD-Q1031-88311, Revision 0, utilized the seismic input contained in BFN-50-C-7104-7, attachment 8.
Calculation CD-Q1031-88313, Revision 0, is a support calculation which utilized the load generated from CD-Q1031-88311, Revision O.
In the anchor bolt calculation contained in CD-Q1031-88313 Revision 0, pry'ng action was not explicitly considered.
This is because the baseplate was judged to be relatively rigid and there existed ample safety margin in the i
i anchor bolt design.
In July 1988, because of other design changes, this calculation was revised to account for a thirty percent increase in design loads.
During this revision, the baseplate and anchor bolts were reanalyzed with the Baseplate II computer program which considered prying action.
The anchor bolts were still found acceptable with the increased loads.
This later analysis validates the judgment used in not considering prying action in l
revision 0 of the calculation.
i t
i I
s
Page 5 of 20 Concern 7.
Calculations 481004-MS2-75-R5 and -12 indicate that certain steel members are overstressed.
The ovarstressing was alleviated by a reanalysis of the STRUDL computer model in which the anchor at Node 23 was deleted.
TVA should review this practice to ascertain the adequacy of the affected steel members.
TVA Response Calculation 48N1004-MS2-75-R12 identifies the overstressing of certain components of the miscellaneous steel support framing for Core Spray supports R-12. H-27, and H-28.
A modification was developed in the calculation to resolve the overstressed conditions and is analyzed in the calculation by incorporating changes to the STRUDL model.
The modification is shown on drawing 48H1004-2 Revision 0 and field implementation is complete.
Calculation 48N1004-MS2-75-R5 identifies the overstressing of certain components of the miscellaneous steel support framing for Core Spray suppo ts R-5, H-7, and H-8.
The overstressed condition is the result of the loadiny from a 10" diameter pipe anchor for the Containment Inerting System which is attached to the miscellaneous steel frame.
The overstressed condition was resolved by assuming that the pipe anchor would be removed.
This unverified assumption is documented in the calculation.
Engineering determined that removal of this pipe anchor was a better solution to the overstressed condition as compared to implementing major modifications to the miscellaneous steel frame. Work is currently under way to replace or modify the pipe anchor to allow removal of this unverified assumption from the calculation.
Additional concerns in the Civil Structural area from Section 4.1.2.2 of report Concern A.
The team noted that TVA has committed to requalify the safety-relt.ted buried pipe at Browns Ferry as part of the D8VP.
Buried Class I piping is designated as essential calculation type 0.14 in the civil engineering master calculation list (Reference 4 in Section 4.1.1 of this renort).
l TVA should revise Section 4.4 of Design Criteria Document BFN-50 -7103 to incorporate more explicit design criteria before implementing this commi tment.
TVA Response A Design Input Memorandum (OIM) to design criteria BFN-50-C-7103 will be developed to provide more explicit requirements for the design and evaluation
{
of buried structures / features.
The DIM will be issued by 11/30/88.
\\
Page 6 of 20 j
MECHANICAL AND NUCLEAR SYSTEMS Concern 8.
Some of the mechanical calculations prepared by Ebasco that were reviewed did not list specific criteria, did not draw any conclusions and, in the case of check valve flow, did not utilize conservative values.
In addition, the calculations were not prepared in accordance with the governing design procedura.
TVA Response The calculations (HD-Q2023-87123 and MD-Q2023-87298) that were reviewed have been revised in accordance with the governing design procedures to list specific criteria, provide conclusions, and incorporate conservative values.
A calculation improvement program has been implemented for contractor generated calculations, the details of which are further discussed in our response to concern number 11.
Concern 9.
TVA's contract with Ebasco requires that TVA review the first calculation of each type and provide its comments to Ebasco so that Ebasco can incorporate the comments in the calculation packages.
The NRC team found that TVA is not following up to ensure that Ebasco is incorporating the comments after the comments are transmitted to Ebasco, thus the comments may or may not be incorporated by Ebasco.
TVA Response Each one of the specific types of calculations generated by Ebasco undergoes a detailed technical adequacy review by a TVA reviewer.
The comments of the TVA reviewer become a QA record.
The DBVP systems engineers have been instructed to perform an acceptance review of all calculations in accordance with the memorandum, Acceptance Criteria for Browns Ferry Nuclear Plant Calculations.
As indicated in this memorandum, the 08VP systems engineers are instructed to obtain the original reviewer's comments from the detailed technical adequacy review and verify that the comments have been incorporated in the calculation.
This acceptance review process will enst*e that the comments are adequately resolved prior to the issuance of these calculations.
Open items from Inspection Report 87-36 Concern B.
Some design requirements contained in FSAR Appendix C have not been included in the commitments / requirements (C/R) data base listing.
TVA stated that it intends to include the information in Appendix C in the C/R data base; TVA started this work just before the second NRC inspection ended.
If this work is satisfactorily completed, the TVA revised response would be acceptable.
~
~
.c
~.
Page 7 of 20 TVA Response The commitments / requirements (C/R) data base has been updated to include
(
design requirements contained in FSAR Appendix C.
A total of 15 new C/Rs were generated from the review and were added to the data base.
NUCLEAR SYSTEMS Open_ item from Inspection Report 87-36 Concern C.
The systems mode requirements for the DBVP Restart Plan identified i
various events and actions defined in the safe shutdown analysts (53A).
A review of this document revealed that the containment purge valve (CPV) operational capability was assigned as a Phase 2 priority item (post-restart of unit. 2).
The team felt that this should be assigned as i
a Phase 1 priority item because of the importance of purging the containment atmosphere late in the accident sequence to expel the bulldup of hydrogen gas in the containment and mitigate the possible consequences of a hydrogen explosion inside containment.
TVA Response Although this item was closed in audit report 88-07, the basis for closing this item was the response previously made and that TVA would functionally test this valve to open against 30 psig containment pressure.
The previous i
response described TVA's position for this valve and felt sufficient information was provided to preclude including the CPV operational capability during Phase 1.
However, during the April audit, additional Information was requested to justify our position.
Vendor procurement data and walkdown data was provided that Indicated the installed valves were capable of opening against the differential pressure.
Therefore, it is TVA's position that the Information supplied during the audit in combination with the previous response justifies the CPV as being Phase II and no testing of the valves to open will be performed.
This position has been discussed with P. Hearn of your staff.
i Concern i
- 10. Nuclear systems essential calculations have not yet been identifled.
Such an identification is required to satisfy the objectives of OBVP and should be done as soon as possible.
i TVA Response The Nuclear Technical Branch (NTB) DSVP calculations have now been reviewed and the essential DBVP calculations have been identified.
i r
.,r~
'- - - ~ -
- - - - - - ~ - - =
Page 8 of 20 Concern
- 11. The Engineering Assurance (EA) Group of the DBVP has not reviewed any nuclear calculations even though the calculation process is well under i
way.
l TVA Response At the time of the NRC audit in April 1988, most of the nuclear calculations available for review by the Engineering Assurance Oversight Review Team t
(EA-ORT) were to satisfy DBVP needs related to identifying scope of the program. Design change calculations were in the process of being technically reviewed and/or regenerated.
EA was scheduled to review representative samples of these design change calculations af ter the process was further along.
Ten items such as the safe shutdown analysis and a system requirement calculation, to mitigate FSAR Chapter 14 accidents, were reviewed by EA-ORT previous to the audit and should have been, but where not, made available to l
NRC.
In July, EA performed a programmatic audit during which 3 nuclear design change calculations were reviewed and no technical deficiencies were identified.
Further technical audits of nuclear calculations are scheduled during the last quarter of 1950.
In addition, in August 1988. EA requested three major engineering contractors (Bechtel, Ebu co and SWEC) to develop calculation improvement programs.
Such programs are now in place and include:
(a) Independent peer reviews of l
existing calculations, (b) QA/QC survelliances utilizing checklists /guldelines covering "lessons learned" based on previously identified problems, (c) i l
feedback to line organizations from (a) and (b) above. (d) strengthening the training of personnel on calculation procedural requirements, and (e) corrective action of any problems identified during the reviews and/or
]
surveillances.
EA has regularly monitored the effectiveness of the calculation improvement process at BfN since September 1988.
s Concern
- 12. The NRC team reviewed a number of nuclear calculation packages and found i
that some assumptions were not clearly stated and that others lacked proper reference to appendices ano attachments.
I 2
L TVA Response Unverified assumptions in NTS 08VP calculations will be addressed and
[
dispositioned, as required, by the Volume 3 calculation review commitment, by l
restart of unit 2.
l Concern
- 13. Because certain types of n xiear calculations are configuration dependent, TVA needs to v iew those calculations after the plant configuration is reestabisshed.
l 1
.t
l
~*
Page 9 of 20 TVA Response The NE Engineering Assurance Branch will perform a plant configuration review on a sample of the NTB calculations before unit 2 restart to ensure that the calculated results represents the DBVP plant configuration.
Concern 14.
Nuclear Calculation Procedure NEP-3.1 does not require that a table of contents be provided for calculation packages.
This procedure should be revised to require a table of contents.
TVA Response Nuclear Calculation Procedure NEP-3.1 is in the process of being revised to require that a table of contents be provided for calculation packages.
The procedure e.hange notice to NEP 1., will be issued by November 15, 1988.
Concern 15.
In some instances, inputs to computer runs and computer codes used in the nuclear calculations were not always stated in the calculation packages.
TVA should include this information in the applicable calculation packages.
TVA Response NTB practice in the past has been to flie the software printouts separately from the calculation and reference the location of the file in the abstract portion of the calculation. Now. Nuclear Engineering Procedure (NEP) 3.1, Calculations, issued in July of 1986, instructs the calculation preparer to eliminate such problems. Section 4.1.2 states that the calculation preparer:
"Ensures that for all computer programs used to perform computations or analyses:
The program has been verifled and documented in accordance with NEP a.
3.8.
b.
The sof tware version, computer input, and computer output are docuented, retrievable, and referenced in the calculation document."
Moreover. Section 4.1.6.2 in this procedure speelfles that the inttlating manager "ensures that the calculations and supporting documentation (including computer input and output) are issued as in NEP 3.1."
Applicable personnel are tralned in the use and application of NEP 3.1.
A review of the references ilsted in section 4.2.2.1 of the audit report indicate that all calculations that were reviewed by NRC were prepared prior to the issuance of NEP 3.1 except BFN-APS3-011, Tnts calculation has been reviewed and no problems related to this concern were found.
t
i Page 10 of 20 During the calculation review program, if these types of documentation problems are identified, they would only be corrected during the next revision of the calculation.
Therefore, TVA considers that this concern has been addressed with the issuance of the NEP 3.1.
Concern
- 16. Calculation TI-ANL-69 identifled a possible safety concern.
The reactor core isolation cooling (RCIC) turbine electre.lc overspeed trip (at 1107 rated speed) will be actuated if flow is allowed in the pump discharge mini-flow line.
Unless this trip is successfully overridden, no allowance can be made for cooling water provided by the RCIC systems.
TVA must review this problem to ensure that the turbine trip can be successfully overridden.
Additional details related to this concern from Section 4.2.2.2.2 of report In addition, many NTB calculations are required by other engineering groups within TVA.
It was not possible to determine if the required information or calculation results had been transmitted to the pertinent group as required o* needed. Documentation should be available to show that follow-on actions to resolve problems identifled by NTB calculations were accomplished as required. Also, documentation should be in place to show that the NTB calculations used as inputs by other engineering groups are the most current and correct versions.
Letters of transmittal or other forms of documentation for the NTB calculations were not provided.
TVA Response Calculation TI-ANL-69 was issued in April 1982.
At that time less formal means were utilized to convey internally generated design information.
NEP 3.3, Internal Interface Control, dated July 1,1986, established requirements and methods to control internal design interfaces and for requesting or conveying design information across disciplines.
Such practices ensurs that internal design information is communiceted in an effective and timely manner.
The fact that failure of the mini-flow valve to close may result in a RCIC turbine trip is noi: a safety concern because RCIC is not a single failure proof systes. Other valves and components can disable RCIC in the event of their single failure.
The HPCI and ADS in combination with the low pressure core cooling systems are also available in the event of such failure.
An NT8 review of this calculation has indicated that it was performed to provide inputs needed for a Probabilistic Risk $ssessment (PRA) being performed on the Browns Ferry Nuclear Plant.
Additionally, during the recently completed review of essential NTB OSVP and Equipment Qualification calculations, TI-ANL-69 was determined to be a desirable calculation, a classification created for important calculations that have no safety-related aspects.
Page 11 of 20 Concern t
- 17. Procedures EN DES-EP 3.23 and NEP-3.8 do not require verification of compatibility of public domain software with the system at Browns Ferry.
Some computer code verification should be required to ensure software compatibility with the system and consistency of results.
Additional details related to this concern from Section 4.2.2.2.2 of report f
Moreover, prior to May 1979, there was no procedure in place to document, control and verify computer codes used in TVA engineering calculations.
This raises questions concerning the degree of technical review that was done on i
TVA engineering calculations utilizing computer codet before May 1979.
]
TVA Response The g(verning procedure for computer software control is NEP-3.8 which is in the pr:,:ess of being revised to require verification of compnibility of pubile domain software and consistency of results with the computer system at 4
BFN.
The procedure change notice (PCN) to NEP 3.8 will be issued by November 15, 1988. A survey to identify each application of public domain software has been completed. Certification for each softwarc version uttilzed in DBVP and equipment qualification calculations has been requested from the i
software owner.
This will be completed, as part of the Volume 3 calculation review comitment, by unit 2 restart, t
1, l
Additional concerns in the Nuclear Systems area from Section 4.2.2.2.2 of report j
Concern l
D.
There were two copies of Calculation TI-764 that were not identical; neither was a copy of the other, a
i
)
i i
TVA Response
(,
This calculation was issued in January 1981.
The problem originated because i
i the calculation preparer did not remove the revition zero coversheet which had 1
been marked up with changes the typist needed to prepare the revision 1
}
i coversheet. The quality of the revision 1 calculation was nort degraded by i
having a superfluous coversheet.
This problem was considered to be an l
1 isolated case.
t l
The NTS calculation classification effort has determined that TI-764 is a i
"flie only" calculation tt.at is retained only for record purposes.
The
(
augmented procedure, NEP 3.1, Calculations, issued in July 1986 provides 4
additional measures that will preclude such an occurrence.
L Concern t
l E.
One calculation had pages that were not signed, had no page numbers, and
[
did not indicate which calculation that they were part of.
j i
l
Page 12 of 20 t
TVA Response The calculation found to have such documentation defects during the audit was completed before the July 1986 issuance of NEP 3.1.
Such a practice.ls now unlikely because of the augmented procedures for calculation preparation, review, and approval that are part of this procedure.
These problems Identified during the audit were documentation probiems that did not degrade i
the technical content of the calculation.
L Concern l
F.
Calculation TI-ANL-16 was mostly illegible.
TVA Response No further problems of this kind have been identified during the recently completed NTB calculation review.
Therefore, this kind of problem is considered an isolated case rather than a generic NTB calculation problem.
During the review of NTB calculations TI-ANL-16 was determined to be a f
i, superseded calculation that is retained for record purposes only.
Nuclear Engineering Procedure 3.1 provides adequate Instructions and reviews to preclude any further such problems.
4 1
El.ECTRICAL SYSTEMS l
Open items from Inspection Report 87-36 1
Concern l
G.
Apparently, TVA has not considered end-of-life conaltions and approved maintenance actions in design criteria and test requirements documents i
for vital de power systems.
i j
TVA Revised Response 1
l The finding addresses two separate concerns:
I r
i 1.
End of Ilfe conditions for the batterles and their relationship to the l
1 acceptance criterla in the Test Requirements Documents (TRO), the calculations, and the Design Celterla Documents (OCD).
i i
2.
Maintenance actions and their effect on the TRO, the calculations, and the DCD. Cell removal is a specific concern.
a
)
The DC0 specifies the following.
1.
The design criteria for the system specifies that 250-V de system must maintain the voltage between 210-V de for battery end of discharge and 277.8,1.8-V de for battery equalizing charge, 3
j 2.
Battery sizing is to be performed in accordance with IEEE Standard 485-1983, which requires that normal cell deterioration be i
addressed.
Page 13 of 20 3.
Periodic battery performance discharge tests of the unit batteries and shutdown batteries are made in accordance with surveillance instructions to determine the condition of the batteries. Analysis of the tests determines when batteries should be replaced.
There are no preapproved maintenance actions affecting battery cell configuration.
Strapping out of a cell requires a Temporary Alterdien Control Form (TACF) or design change.
Either of these processes.
require an engineering evaluation prior to implementation.
Additional concerns from Section 4.3.2.2 of report Concern H.
Questionable Methodology of Containment Electrical Penetration Calculations Identified by TVA Engineering Assurance Audit The calculations use the approach given in ICEA Standard P32-382-1969 for generally predicting the short circuit thermal capacity of insulated cables, instead of the penetration manufacturer's short circuit test data.
The calculations do not sufficiently demonstrate that the general ICEA formula gives conservative results compared with actual test data, as TVA and Bechtel maintained when the team called the issue to their attention.
The criterion for acceptable penetration protection adopted in the calculations is that overcurrents shall be interrupted, before thermal damage to the penetration, by the first-line circuit breaker or fuse, rather than the backup protective device as recommended by NRC Regulatory Guide (RG) 1.63.
In its initial oral response. TVA pointed out that Browns Ferry has never been committed to RG 1.63, which was issued after the plant was built.
However, containment penetration integrity is critical to the "defense-in-denth" principle, and experlince at other plants (e.9., Sequoyah) suggests that adequate protection by backup devices can be obtained (without hardware modifications) throughout most, if not all, of the range of possible overload and short circuit currents by appropriate protective relay and circuit breaker trip settings.
Therefore, TVA should ensure that penetration protection is a high-priority objective of the protective device coordination analysis now in progress.
TVA Response Although BfM is not committed to RG 1.63, TVA has ensured that containment penetration integrity is being maintained.
The penetration protective device coordination analyses have been completed for every circuit which passes through a penetration.
This analysis used accepted engineering practices to assure that the thermal capability of the penetration was adequately protected, thereby ensuring containment penetration integrity.
Eatremely conservative assumptions were utilized to assure adequate safety margins.
Page 14 of 20 The method employed calculated the maximum allowable short circuit thermal I
capacity for different conductor sizes, using ICEA P32-383-1969.
This formula calculates the current withstand capability as a function of the Ist value i
and the temperature differential from the initial conductor temperature to the l
maximum allowable final temperature. This formula wss very conservatively applied to assure adequate margin by assuming the final temperatore was 250'C as compared to the penetration insulation capability of 400*C.
l The electrical penetrations were reviewed for short circuit and short time l
fault currents, and for the total heat load in each penetration module to assure that the nozzle / concrete interface temperature and the thermal l
capability of the individual modules are viot exceeded.
It should be noted that in all cases, even with the conservative assumptions, the calc.ulated heat load of the penetration modules is well below the rated maximum watts per foot values furnished by the penetration manufacturers.
To calculate the maximum heat load, the continuous current was assumed equal to the rated trip value of i
the protective device for each conductor, which is very conservative.
Additionally, an estimated current was assumed for the spare conductors and all conductors were assumed energized.
Furthermore, it should be noted that in all cases, the penetration conductors are of equal or larger wire size than the associated field cables.
Therefore, continuous overload capability of the conductors was not considered a concern.
Concern 18.
The NRC team's review of the sizing methodology for the batterles for the vital de power systems revealed that the definition of load profiles in the duty cycles was not sufficiently conservative.
TVA must revise the methodology to include conservative operating time and ampere values for the active loads considered in the duty cycle.
TVA Response The battery sizing calculations are currently being revised to better reflect the actual plant loading conditions.
In making these revisions. TVA will evaluate the methodology used for sizing the batteries and take into consideration the comments made by NRC concerning traceability of the documentation of various data used in te load analysis.
The revisions to these calculations will be completed, as part of the Volume 3 calculation review commitment, by testart of unit 2.
Concern
- 19. The NRC team's review of EA Action Items E-034 and E-035 revealed that the EA reviewers had not reviewed precursor calculations (inputs to calculations E-034 and E-035); thus, their validity could not be verifled.
TVA should revise the EA sampling program to include this type of verification.
Page 15 of 20 i
TVA cesponse The EA Oversight Review Team (EA-ORT) did not randomly select calculations to i
review as may have been inadvertently conveyed to NRC during the audit.
EA-ORT selected representative samples of different types of calculations to verify technical adequacy.
While performing the technical review, the l
1 references to other calculations used as design input are vertfled as i
correctly referenced. Accuracy of data transferred into the successor calculation and correct usage of the predecessor information is also verified.
TVA does perform, in some cases, a vertical slice review, i.e.,
l review each calculation input for technical adequacy in a chain of referenced i
predecessor / successor calculations. However, since this was a representative L
calculation review, EA considers that the technical review methodology used I
was acceptable in this instance.
Concern
- 20. The NRC team's review of electrical calculations prepared by Bechtel found that the calculation packages did not include references and assumptions.
TVA must revise these electrical calculations to include both.
TVA ResDonse TVA has discussed the above concern with Bechtel who will address references and assumptions in the calculation revisions which are now in progress.
Additionally, TVA is performing a review of the Bechtel calculations to insure their adequacy.
The review and revisions of these calculations will be completed, as part of the Volume 3 calculation review commitment, by restart of unit 2.
INSTRUMENTATION AND CONTROL SYSTEMS Open items from Inspection Report 87-36 Concern I.
A comprehensive systemwide walkdown or functional test of ILC systems was lacking.
TVA's response did not provide all of the required information.
This remains an open item pending TVA's new definition of baseline configuration.
TVA Response See response to concern J.
Page 16 of 20 i
Concern i
J.
Establishment of a true configuration baseline has to be established.
l TVA clarlfled the term "baseline configuration" and stated that the l
Intent in the I&C area had been to establish the TVA baseline functional configuration.
TVA plans to submit an amendmelt to this response.
This remains an open item pending the NRC team's review of this amended response.
TVA Response 1
In addition to establishment of baseline functional configuration for I&C components by testing, a physical walkdown was performed for the I&C control diagrams and instrument tabulations.
This walkdown provides nameplate data verification for I&C components. Additionally, for I&C components which have a setpoint, or for indicators h. '.. are required for restart, calculations are performed to determine the demon.trated accuracy of the instruments, and the suitability of the setpoints.
Each ISC component involved in these calculations have the appropriate data field verified prior to issuance of the calculation.
i Concern K.
Double cross-wiring because of the lack of a comprehensive configuration
(
check is likely.
TVA stated that the probability for double cross-wiring is extremely
(
low. However, the NRC team noted that if double cross-wiring did occur, l
a large number of physical wirings, termir."lons, and components would be l
Involved and an adequate probability risk analysis would have to be l
l performed.
This is still an open item pending further TVA action, l
j i
TVA Response
{
l Tne scope of the OBVP is to provide functional verification of the schematic and elementary diagrams through functional testing.
Each system is reviewed
{
to ensure that the testing provides this functional verification and is L
adequate to prove safe shutdown capability of the plant.
Phase I of the OBVP i
is to establish the functional configuration of the plant.
BFEP PI 86-03, i
Preparation and Control of Engineering Change Notices Packages, is the
}
procedure to control future modifications to the plant.
This procedure requires a physical walkdown verification of the system, or portion of j
systems, affected by the changes before the changes are implemented.
BFN has been an operating plant since 1974 with surveillance tests and instructions carried out periodically.
Problems caused from double cross-wiring probably would have surfaced during the operation and testing of equipment. Also, the l
third cycle of trend analysis of the CAQs revealed no deficiencies caused from 4
j possible double cross-wiring, j
l b
i J
- I
l Pa 17 of 20 Concern i
21.
Section 3.8 of the Emergency Equipment Cooling Hater (EECH) Design i
Criteria Document 8FN-50-7067 states that In-titute of Electrical and Electronic Engineers (IEEE) S?andard 279-1971 is appilcable to the system flow-measuring instrument.
This is incorrect because the EECH flow a
instrument monitors a Regulatory Guide (AG) 1.97 Type-O variable, which requires only Category 3 (commercial) quality rather than Class IE l
quality equipment. Moreover, the appilcability of this standard to othe -
l portions of the EECH System was not stated.
TVA should make these L
determinations and correct the design criteria.
i TVA Response EECW Restart Design Criteria Document BFN-50-7067 R1, Section 3.8(2),
Regulatory Requirements, Codes. and Standards, states that the applicability l
of IEEE Standard 279-1971 is limited to instruments for cooling water flow Indication.
TVA agrees that this standard does.iot apply to the EECW System i
flow-measuring instrument.
This particular IEEE standard should apply to all i
redundant safety protective instruments.
EECH Restart Design Criteria
[
BFN-50-7067 R1, Section 3.8(2), has been revised by Design Input Memorandum
[
(DIM) DIM-BFN-50-7067-2 to clarify the applicability of IEEE Standard 279-1971.
Concern
- 22. A pressure transmitter (2-PT-1-72) was downgraded to a nonsafety-related status.
This downgrading tsolates IEEE Standard 279-1971 and NRC Branch t
Technical Position ICSB-26, "Requirements for Reactor Protection System Anticipatory Trips," (NUREG-0800).
TVA should reexamine the other 69 l
Unit 2 instruments to ensure that no other problems exist.
[
TVA Respny Pressure tralsmitters 2-PT-1-72, -76, -82, and -86 monitor main steam Ilne i
pressure at the inlet to the main turbine.
The transmitters and their associated switches are to detect abnormal transient events suci as l
inadvertent ogning of the turbine bypass valves when the reactar is at pressure.
Sudden depressurization is the basis for closure of trie main stear, isolation valves (MSIV) to prevent undesirable transients on the rcactor internals. The four transmitters and switches operate in 1-out-of-2 twice fallsafe logic.
This design corresponds to the standard GE-BHR design of the Browns Ferry vintage.
This design was estabitsbed prior to issue.of IEEE Standard 279-1971.
However, the design was avaluated against and does comply with the Intent of IEEE Standard 279-1971.
This evaluation is documented in General Electric NE00-10139 dated June 1970.
I
Pagt 18 of 20 Branch Technical Position ICSB-26 addresses reactor protection system (RPS)
"anticipatory trips" and establishes the position that all inputs to the RPS must moet the requirements of IEEE Standard 279-1971; 1.e, the effects of credible faults or failures in these anticipatory trip functions must not be capable of propagating back to the RPS and degrade the RPS performance or reliability.
The pressure transmitters provide input to the primary containinent isolation system logic cabinets via the analog trip units (ATU).
The pressure transmitters are the only devices within this instrument loop that have been downgraded to a nonsafety-related status by the quality information release.
The ATUs are a safety-rela'ad interface and provide quallfled isolation to meet the intent of IEEE Standard 279-1971 The other 69 instruments mentioned in this concern have been evaluated by the Design Baseline Verification Program (OBVP) as part of the effort to identify "essential" calculations.
None of the 69 instruments were identified as requ; ring instrument accuracy calculations prior to unit 2 restart.
These instruments are not required to mitigate design basis accidents, abnormal transients, or special events.
Concern 23.
The NRC team reviewed 10 Division of Nuclear T,oleering (ONE) calculations and noted that four were limited to verification of the retrievability of the calculations.
The four calculations did not meet 2
Attachment E of the DBVP which stated that the objectives of Phase 1 of the OBVP are to ensure that the essential calculations are technically adequate.
TVA should ensure that the instrt, mentation and control calculations are techolcally adequate.
TVA Response The above-referenced calculations were originally generated in response to a Sargent & Lundy finding resulting from their audit of TVA's calculation program.
Sargent & Lundy recommended that the results of manufacturer's calculations prepared in support of safety related equipment procurement contracts should be readily retrievable.
These calculations were prepared by the vendors under a TVA reviewed and approved Quality Assurance program.
The technical adequacy of the vendor calculations was established as part of the equipment procurement process.
The retrie, ability calculations were performed as a means of identifying and documenting the existence of the vendor i
calculations.
Concern
- 24. Calculation 110386BDP-1, "Flow Element Orlfice Plates," included an examination of six safety-related and six nonsafety-related installations for straight-line piping lengths. All 12 installations were determined to be satisfactory The conclusion derived from this calculation was the basis for accepting all other nonsafety-related flow orifice installations at Browns Ferry.
The NRC team believes that the sample size of six nonsafety-related flow element installations is not large enough to be the basis for accepting all the nonsafety-related flow orifice installations at Browns Ferry.
d
Page 19 of 20 TVA Response Calculation 110386BDP-1 RO was revised to EO-00000-88303 RI.
The revision removed any reference to nonsafety-related orifice plates.
The calculation now applies only to the six safety-related orifice plates listed in the calculation which represent the entire population of Electrical Engineering Branch purchased safety-related orifice plates.
Therefere, sampling is no longer involved in tMs calculation.
The nonsafety-related ortfice plates are not required to be evaluated before restart n indicated in the OBVP plan.
This meets the DBVP plan commitment for resti t.
C. s ern 25.
The NRC team noted that in Calculation 033087750, "Verification of Sepa-ration Criter ia for Sensing Lines Calculations," the Electrical Design Standard (DS-E18.3.9, Revision'0) attached to the calculation incorrectly emphasized control-to. protection system separation rather than the separation between redundant portions of the ~ protection system.
The correct reference for sensing-line separation is IEEE Standard 279-1971 Section 4.6, "Channel Independence."
In addition, the calculation referenced 10 CFR 50, Appendix A, General Design Criterion (GDC) 24, as the 18-inch tubing separation design criterton.
This reference is incorrect since GDC 24 does not provide quantitative separation criteria.
The NRC team noted that the 18-inch sevaration requirement does exist in a TVA design criteria document for the Enquoyah plant, which would be an appropriate reference for Browns Ferry.
_T_VA Response Design Standard 05-E18.3.9, Revision 0, was generated to define the design requirements for the separation of redundant instrument lines.
It was not intended to emphasize the control-to-protection system separation rather than the separation between redundant portions of the protection system.
The design standard has been revised to clarify the references listed and more clearly define its purpose.
The remaining part of the concern addresses the references Ilsted in the Design Standard 05 E18.3.9, Revision 0.
This design standard was written to be very specific and in the process some of the general references were omitted.
The design standard has been revisec and issued to clarify the references that were used in its generation.
Since this concern involves no change to technical requirements, calculation 033087750, Verification of Separation Criteria for Sensing Lines Calculations, will be revised post-restart, as part of the Volume 3 calculation revitw commitment, to include the revised Otstgn Standard (05-E13.3.9, Revision 1) in lieu of Revision O.
This action wil) resolve the abovre concern.
(
i
=1
Page 20 of 20
=
Concern
- 26. Tha NRC team reviewed Ebasco's setpoint calculation for reactor building closed cooling water (RBCCW) system time delay relays (CD-02070-99069).
The calculation was listed as being essential (i.e., it addresses a safety-related component), but the calculation (pages 1, 5, and 8) stated that RBCCH pumps and RBCCH time celay relays performed no safety function.
This statement is inconsistent.
The time delay relay does perform a safety-related function as the RBCCH system is converted from two-pump to one-pump operation.
The Ebasco calculation should be modified to clarify that postaccident RBCCW flow is not nonsafety-related, but that the electrical controls needed to reconfigure the system are safety-related.
TVA Response Under the present configuration, ths time delay relays do perform a safety-r31ated function in that a misoperation of the time delays could resJ1t in an unanalyzed loading of the diesels.
The calculation will be revised, as part of the Volume 3 calculation review commitment, to make it a safety-re16ted calculation before unit 2 restart.
l r
/
o Summary List o# Commitments 1.
Issue design input memorandum to BFN-50-C-7103 by November 7), 1988, to provide explicit requirements for the desigr and evaluation of buried structures / features.
2.
The NE Assurance Branch will perform a pidnt configuration review on a sample of the NTB calculations before unit 2 restart to ensure that the calculated results represents the DBVP plant configuration.
3.
A procedure change notice to NEP-3.1 will be issued by November 15, 1988 to require that a table of contents be provided for calculation pack 6ges.
4.
A procedure e.hange notice to Nff'-3.8 will be issued by Novemt'er 15, 1988 to require verification of compatibility of public domain software and consistency of results with the computer system at BFN.
The following commitments are addressed by this submittal ar,d in the Nuclear Performance Plan Volume 3 commitment fcr DBVP restart and post-restart calculation review. As such, TVA will track, followup, and insure completion, in accordance with Volume 3 and does not consider these as n9w commitments.
RESTART 5.
Unverifled assumptions in the NTB DBVP calculations will be addressed and dispositioned as required before unit 2 restart.
6.
The battery si:Ing calculations are currently being revised to better reflect the actual plant loading conditicas.
In making these revistor.s, TVA will evaluate the methodology used for sizing the batterir-and take into consideration the comments made by NRC concerning traces
'ity of the documentation of variou, data used in the load analysis.
The revisions to these calculations will be com11ete before restart of unit 2.
7.
TVA has discussed the above concern with Bechtel who will address references and ass. i.Elons in the calculation revisions which are now in progress. Additlo 4 iy, TVA is performing a review of the Bechtel calculations to insure their adequacy.
The review and revisions of these calculations will be complete by restart of unit 2.
8.
The calculation (ED-02070-99069) will be revised to make it a safety-related calculation before unit 2 restart.
9.
Certification for ea;h software version utilized in DBVP and equipment qualification calculations has been requested from the software owner.
This will be completed by unit 2 restart.
POST-RESTART l
10.
Revise calculation 03308775D; Verification of Separation Criterth for Sensing Lines Calculations, to include Design Standard DS-E18.3.9 Revision 1 as an attachment.
This revision will be implemented as a post-restart commitment.
l l
l l
[
._.