ML20235G341
| ML20235G341 | |
| Person / Time | |
|---|---|
| Site: | Sequoyah  |
| Issue date: | 07/02/1987 |
| From: | Gridley R TENNESSEE VALLEY AUTHORITY |
| To: | NRC OFFICE OF ADMINISTRATION & RESOURCES MANAGEMENT (ARM) |
| References | |
| NUDOCS 8707140210 | |
| Download: ML20235G341 (34) | |
Text
p f
e
-TENNESSEE VALLEY AUTHORITY CH ATTANOOGA. TENNESSEE 374o1 SN 157B Lookout Place l-1
' UL 0 21987 J
- U.S. Nuclear Regulatory Commission ATTN:
Document Control Desk I
l Washington, D.C.
20555 Gentlemen:
In the Matter of
)
Docket Nos. 50-327 Tennessee Valley Authority
)
50-328 i
TENNESSEE VALLEY. AUTHORITY'(TVA) - DIVISION OF NUCLEAR ENGINEERING (DNE) l DESIGN CALCULATION EFFORT FOR SEQUOYAH NUCLEAR PLANT (SQN) l On February 13. 1987, an NRC inspection team from the Office of Inspection and L
Enforcement (IE) in Bethesda, Maryland, concluded a two-week' inspection of the l
DNE design calculation efforts. The team inspected the DNE efforts to identify, retrieve, and review for technical adequacy the essential calculations within each engineering branch, as described in a January 20, 1987 letter to NRC.
The team reviewed the individual branch programs as well as specific calculations for technical adequacy. The results are identified i
as 31 observations and are documented in IE Inspection Report Numbers 50-327/87-06 and 50-328/87-06.
l consists of a restatement of the NRC observations, with the TVA response immediately following the respective observation.
Five categories of observation identification were established, relating to the overall DNE q
programmatic effort and to the four DNE engineering branches. consists of new commitments within enclosure 1, l
If you have any questions concerning this issue, please telephone i
Beth L. Hall, of the Sequoyah Site Licensing Staff, at (615) 870-7459.
Very truly yours, i
TENNESSEE VALLEY AUTHORITY 1M@t
. Gridley, Director Nuclear Safety and Licensing Enclosures cc:
see page 2 8707140210 870702 PDR ADDCK 05000327 PDR g&*l G
1 An Equal Opportunity Employer I
l
4.
U.S. Nuclear Regulatory Commission JUL 0 21987 cc (Enclosures):
Mr. G. G. Zech, Assistant Director for Inspection Programs Office of Special Projects U.S. Nuclear Regulatory Commission 101 Marietta Street, NW, Suite 2900 Atlanta, Georgia 30323 Mr. J. A. Zwolinski, Assistant Director for Projects Division of TVA Projects Office of Special Projects U.S. Nuclear Regulatory Commission 4350 East West Highway EWW 322 Bethesda, Maryland 20814 4
Sequoyah Resident Inspector Sequoyah Nuclear Plant 2600 Igou Ferry Road Soddy Daisy, Tennessee 37379 i
f 1
ENCLOSURE 1 NRC OBSERVATIONS AND TVA RESPONSES The following NRC observations are written as they appear in IE Inspection Report Numbers 50-327/87-06 and 50-328/87-06. The TVA response follows each observation, with the exception of CEB-2 through CEB-6 which apply to a common subject.
CENERAL OBSERVATIONS Observation No. CEN Substantiated Condition for a CAQ Nuclear Engineering Procedure (NEP) 9.1, Corrective Actions, revised 7-1-86, defines the controlled system within the DNE to document, evaluate and resolve l
conditions adverse to quality (CAQs) relating to all engineering work within DNE.
The NRC had previously reviewed implementation of the corrective action program within DNE/0E carried out under the provisions of OEP-17, which pre-ceded NEP 9.1.
NEP-9.1 requires the documentation of "...any condition which renders an item unacceptable to perform its required function or creates uncertainty concerning its ability to meet design requirements..." The procedure also requires a test for significance; those items determined to be significant are documented via significant condition reports (SCRs), all other CAQs are documented via problem identification reports (PIRs).
The team was informed that NEP-9.1 was in the process of revision to agree with the recently revised corporate QA procedure for corrective actions, NQAM Part I, Section 2.16, revision dated 1-4-87.
Although the draft NEP-9.1 was not available for review, TVA informed the team that the draft was the same as i
the NQAM relative to CAQs. Under the new procedures, all CAQs are to be documented as conditions adverse to quality reports (CAQRs); SCRs and PIRs will no longer be generated within DNE.
Section 4.2 of NQAM Part I, Section 2.6, defines CAQs as (emphasis added):
" Adverse conditions include nonconforming material, parts or components; failures, malfunctions; deficiencies; deviations, hardware problems involving noncompliance with licensing commitments, specifications, or drawing requirements; abnormal occurrences; and nonhardware problems such as failure to comply with the operating license, technical specifications, licensing commitments, procedures, instructions, or regulations.
Unsubstantiated conditions are not defined as CAQs."
The team was concerned with the lack of inclusion of unsubstantiated conditions within the set of CAQs. The licensee does not appear to have an alternative controlled system in place to identify and resolve conditions which create uncertainties regarding an item's ability to perform design functions.
Such situations could be interpreted as unsubstantiated during identification and reviews. The team was concerned that the net effect of the procedure change would be to eliminate a set of CAQs which had previously been identified and resolved by SCRs and PIRs, without establishing an alternative method for identification, tracking and resolution.
TVA Respoase A method currently exists for documenting unsubstantiated conditions.
NEP-9.1, Revision 1, issued on February 20, 1987, provides for implementation of a modified PIR system within the Division of Nuclear Engineering (DNE) to document problems and potential problems that are not CAQs. The issued l
version of Revision 1 differs in this aspect from the draft version discussed with the NRC team, which did not contain a provision for generating PIRs.
l This provision is currently implemented by DNE at SQN by SQN Engineering Procedure (SQEP)-61 Revision 0, " Handling of Condition Adverse to Quality Reports (CAQRs) and Problem Identification Reports (pIRs)."
l Observation No. GEN CAQ Operability Determinations The team was informed that NEP-9.1, Corrective Actions, was in the process of revision to agree with the recently revised corporate QA procedure for corrective actions, NQAM part I, Section 2.16, revision dated 1-4-87.
Although the draft NEP-9.1 was not available for review, TVA informed the team that the draft was the same as the NQAM relative to CAQs.
Under the new procedures, all CAQs are to be documented by condition adverse to quality reports (CAQRs); SCRs and PIRs will no longer be generated within DNE.
l The team noted that the new procedure requires the management reviewer of the organization initiating a CAQR to assess the condition for potential impact on operability (procedure section 5.2.2).
This operability assessment is to be performed in accordance with procedure Attachment 5, Cuidelines for Potential Operability Determinations. This requires that the component's operability be determined by its ability to perform its safety-rela'ted function required by the technical specification rather taan its design-related function.
The team is concerned that restricting the operability assessment to a confirmed tech-nical specification non-compliance inappropriately disregards design based operability requirements that are included in such documents as the FSAR and design criteria.
For example, on some nuclear power plants, there are no specific technical specification operability requirements for room coolers; however, removing these from service could compromise the environmental quali-l fication of safety-related equipment.
TVA Response The Nuclear Quality Assurance Manual (NQAM), Part I, Section 2.16. " Corrective Action," attachment 5, has been revised to address the NRC team's concern relating to guidelines for determining impact on operability.
Revision 3 to the subject section of the NQAM contains the following paragraph, which replaces the third paragraph of attachment 5 contained in Revision 2:
Exceeding design criteria may or may not cause the affected equipment to be inoperable. Design-related deficiencies are to be investigated through calculations, evaluations, communication with vendors or other means to determine whether the deficiency renders the affected equipmenc inoperable.
MEB OBSERVATIONS l
Observation No. MEB MEB Design Calculation Scope of Review Based upon the description of the MEB calculation review sample size provided in memorandum B44870206043 and discussions with DNE personnel, the team recommended the following enhancements to ensure a representative review:
A.
A 100% review of related essential calculations for all calculations determined by DNE calculation review program to be unacceptable. At the time of the inspection two generic areas had been determined to be unacceptable, viz. HVAC heating / cooling load determination calculations and off-design condition calculations.
B.
Calculations produced by all four MEB sections would be sampled.
C.
Calculations dated later than January 1986 would be sampled since review of one calculation by NRC dated later than January 1986 indicated that even new calculations should not be assumed to be correct (see MEB observation No. 6).
D.
Calculations supporting the DBVP would be expanded to include seven of the fifteen calculations within the DBVP scope.
These enhancements were recommended to the MEB Chief Mechanical Engineer from his staff in memorandum B44870211026.
TVA Response All actions indicated in this comment have been completeJ. More specifically:
A.
One hundred percent of the two generic areas were, in fact, reviewed.
In the heating, ventilating, and air conditioning (HVAC) area, each of the seven additional calculations that were reviewed was found to be unacceptable. The seven calculations were then redone and at the same time divided into unique calculations, resulting in a total of 23 calculations.
Of the 23 calculations, eight CAQRs were written to identify individual problems in each deficient calculation (SQT870664-SQT870671).
The corrective actions will be identified, and the CAQRs will be dispositioned in accordance with the IVA CAQ process.
l In the area of off-design calculated conditions, 11 calculations were reviewed; and all 11 were either acceptable or acceptable with minor deficiencies.
l B.
Calculations produced by all four sections were sampled. The number of calculations produced by each area and the number reviewed are tabulated in the following paragraphs:
Hast Cyclo Group:
37 were reviewad of 155 total produced (24 percent).
The results were that three were unacceptable; all attributabic to off-design conditions. The remaining were acceptable or acceptable with minor deficiencies.
Mechanical Equipment:
One.was reviewed of 26 produced (four percent).
It was acceptable.
HVAC, Fire Protection:
34 were reviewed of 134 produced (25 percent).
The results were that all of the cooling load calculations were unacceptable (i.e., 10).
The remaining were acceptable or acceptable with minor deficiencies.
SQN Project:
Six were reviewed of 65 produced (nine percent). All of these were acceptable.
C.
Calculations dated after January 1986 were included in the sample, six from a total of 77.
All were either acceptable or acceptable with minor deficiencies.
D.
Seven of the calculations supporting the Design Baseline and Verification Program (DBVP) were reviewed, and all were either acceptable or
{
acceptable with minor deficiencies.
{
Observation No. MEB 2 - SI Pump Miniflow Rate In reviewing the der <ign interface responsibility between MEB and Westinghouse the team noted that the design criteria for the safety injection (SI) system (B05860805507) required the SI pump minimum recirculation flow (miniflow) to be 40 GPM. Whereas the miniflow orifice drawing (No. RMF-46368 Rev. 0) and TVA correspondence with Westinghouse documented the miniflow to be 30 GPH.
MEB acknowledged after verification with Westinghouse that the design criteria was in error and will be updated to reflect a miniflow rate of 30 GPM.
TVA Response The Mechanical Engineering Branch (MEB) concurs with this observation.
Westinghouse Electric Corporation has agreed to correct the design criteria.
MEB has issued PIR SQNMEB8798 to correct this problem.
Observation No. MEB Water Hammer The design interface between the mechanical systems (MEB) and the pipe stress analysis discipline (CEB) was tested by reviewing the containment spray operational modes calculation (B25861223300). The team noted that no mention was made of water hammer effects. Later it was discovered that water hammer effects (i.e., forcing functions) are transmitted to CEB by NEB rather than via MEB.
Subsequently, CEB verified that no documentation was available which demonstrated that water hammer effects had been evaluated for the containment ___
_J
spray system which is in TVA's design responsibility.
After further evaluation, CEB performed an informal calculation which demonstrated that the effect of water hammer on the containment spray system is insignificant.
CEB should formally update their stress analysis to document that water hammer effects are insignificant and the basis for that determination.
While reviewing the broader issue of systems that had been evaluated for water
. hammer, the team questioned why the feedvater system water hammer analysis had been completed but not issued. CEB needs to justify the reason for not issuing the feedwater water hammer analysis.
TVA Response Transient conditions for the containment spray system (CSS) were considered negligible by the virtue that this systen was not identified to require a water hammer analysis.
During the NRC inspection of DNE calculations in February 1987, TVA performed informal calculations to verify that water hammer loads, as well as dynamic loads, because of filling of the piping system with water, are negligible. These calculatio:as are complete and will be included by. July 1, 1987, as an appendix in the CSS calculation package.
NUREG-3939 is a report on current and re:ommended practices regarding the consideration of dynamic loads in the design of nuclear piping systems and is based on a survey of industry experts. Published in 1984, it had no influence on TVA's past program to qualify nuclear piping systems.
The initial' design of TVA's nuclear plants was done in the late 1960s and 1970s. The recognition of requirements for transient analyses was based on the experience TVA engineers had gained in the design and operation of fossil plants. These engineers identified systems that warranted transient analyses, and additional analyses were performed only when problems arose or new trends developed in the industry. TVA has kept up with induntry trends through participation in code committees, consultation with other utilities, and also through personal services contractors who have performed nost of TVA's early transient analyses.
With respect to the SQN feedwater (FW) line, Westinghouse notified TVA in 1978 that an instantaneous circumferential pire break in the line upstream of the check valve could have a detrimental effect on the FW check valves. TVA followed up with detailed analyses, concluding in 1979 that the check valve design employed at SQN was adequate.
The subject of water hammer analysis for the FW piping did not arise until 1980, the time SQN received its operating license. By 1983, forcing functions were available for the FW piping; but TVA decided not to perform a structural i
response analysis. An evaluation of the pipe whip steel downstream of the l
check valves indicated that there is as much steel in the space as is practical to install, consistent with access for inspection and maintenance.
i Further, this pipe whip steel is shimmed to within one inch or less of the l
pipe. At each restraint, the gap provided was judged to be true minimum l
practical consistent with pipe shakedown, unanticipated thermal expansion, insulation, etc.
l l
l l
1 In summ2ry, ths SQN check valvas and circumferential pipe stress can withstand the pressure spike because of a circumferential break in the FW line. As much pipe restraint steel as has been judged prudent and practical is already in place.
In the unlikely event of this severe accident in the FW system, pipe movement dcwnstream of the check valves will be limited to the pipe support gap. Additional analysis or supports do not appear to be practical or justified.
Observation No. HEB Potential Generic Condition Evaluation l
In reviewing the problem identification reports (PIRs) and significant condition reports (SCRs) generated as a result of the MEB calculation review program, the team noted that none required a potential generic condition evaluation (PGCE). MEB's justification for not requiring a PGCE was that all other TVA nuclear facilities were having similar calculational review programs, Moreover, instructions from the Chief Mechanical Engineer via memorandum B44860225014, stated that a PGCE for SCRs was not required for the identical reason.
Since the calculational review programs are not 100%
comprehensive, but instead are based upon sampling, the aforementioned reason does not ensure that a specific technical issue raised on Sequoyah will also j
be addressed at the other TVA nuclear facilities. Therefore, MEB needs to re-evaluate the need of a PGCE for all PIRs and SCRs generated to date as a part of the calculation review program.
This observation also applies to NEB.
TVA Response HEB concurs with this observation.
MEB is reevaluating the generic applicability of MEB-initiated SQN SCRs and PIRs on other plants.
Instructions were issued by the Chief Mechanical Engineer to perform this review.
A memorandum has been prepared from the Chief Nuclear Engineer to the lead nuclear engineers of other TVA nuclear projects to review similar calculations that have been subject to CAQs on SQN.
These reviews will be conducted consistent with the respective project schedules. Thus, this item has no impact on SQN unit 2 restart.
Observation No. MEB EA's Review of CAQs In discussing a pending change to TVA's proceduro entitled " Corrective Action," NEP-9.1, it was stated that EA will review all CAQs that have potential generic condition evaluation (PGCE) checked "yes."
However, a similar review will not be performed for CAQs with PGCE checked "no."
This i
methodology biases the process to reduce the total member PGCEs generated, rather than verifying if the need for a PGCE has been accurately assessed.
For example, under this system, the MEB SCRs and PIRs that were inappropriately checked "no" for PGCE (See observation MEB-4), would not have had the opportunity to be corrected by EA.
It is recommended that EA be used to verify the adequacy of all CAQs for PGCE, not just CAQs that have been determined to require a PGCE.
TVA Response The development and implementation of the new corrective action program, as defined in NQAM 2.16 and NEP-9.1, intentionally relies on the knowledge and ability of the organization (i.e., responsibic organization most closely associated with and responsible for determining corrective action) to make the initial determination as to whether or not a cited problem has potential generic implications.
For those problems determined to require a generic review, Engineering Assurance (EA) will procedurally performi a further review and appropriately conduct a generic investigation with applicable TVA nuclear projects and organizations.
For those problems determined by the responsible organization to not require a generic review, EA will initially perform audits and surveillance of the cor ective action process by including an inspection attribute to assess the appropriateness of "nongeneric" determinations.
The results of these audits and surveillance will be analyzed to determine the need for further revision to the gercric review portion of the corrective action process.
Observation No. MEB Component Cooling Water System Design pressure In reviewing the calculation which establishes the component cooling water system design pressure (B44861210008), it was noted that the design pressure was established by combining the pump head associated with a LOCA-SI flowrate and the static head between the high and low points in the system.
Consideration of pump shutoff head and surge tank relief valve setpoint was omitted without justification.
If considered, the resulting maximum system pressure would exceed the design pressure by approximately 20 psi. Therefore, MED should assess the impact of the two aforementioned parameters on the design pressure of the component cooling water system as well as their impact on testing requirements for the associated piping and other components.
TVA Response l
Recalculation of the revised system operating pressure, including consideration of the surge tank relief valve setpoint, is complete.
The pump shutoff head is not a consideration because of the system configuration.
The operating pressure remains below the 150 psig design pressure.
Observation No. MEB Identification of Controlling Calculations In reviewing the listing of MEB calculations, the team noted that some calculations addressed the same design attribute, others evaluated off-design conditions, some had been superseded in part by other calculations, and some did not clearly state the purpose of the calculations.
Because of these examples, it is suggested that MEB put in place a system that clearly identifies the controlling calculations for each design attribute. l t
TVA Response MEB concurs with thir observation. Based upon discussions with the NRC team, this observation dealt with having multiple calculations for the same system design parameters. MEB has examined the calculations in the SQN Mechanical Calculation Log to ensure that the categorization (essential, desirable, and superseded) is accurate and complete.
Observation No. MEB Inconsistent Equipment Qualification Temperature Calculation NEB 811007235 was prepared to analyze a deficiency in the reoperation test results for the turbine driven auxiliary feedwater pump (TDAFW) pump room ventilation system and to provide suggestions to reduce the temperature rise.
The discussion and methodology in the calculation uses a temperature rise to 1250F, which the calculation states is the equipment qualification temperature. This is not consistent with the plant's environmental data sheet, 47E235, Sheet 71, that specifies a design temperature of 1040F and a peak temperature of Il00F for the TDAWP pump I
location.
TVA Response MED concurs with the observation.
There is some confusion with the maximum ambient temperature specified for this space.
As a result of our calculation review program, TVA identified an input data deficiency in the cooling load calculations. This deficiency is being tracked by SCR SQNMEB8748, Revision 1.
The corrective action for the SCR required regeneration of the essential cooling load calculations, one of these regenerated calculations is for the TDAFW pump area.
In addition to correcting the calculation, it was necessary to clear up the discrepancy identified by the observation.
pIR SQNMEB8773 was sent to the Nuclear Engineering Branch (NED) for evaluation of temperature limits in this area.
NEB's corrective action established the maximum temperature limits for this 0
space as 110 F.
As part of the corrective action for PIR SQNMEB8773, the FSAR will be changed to reflect the temperature limit rather than the temperature rise now referenced.
NEB calculation NEB-811007-235 has been downgraded to " desirable" by memorandum and is considered a study calculation.
Observation No. MEB Unverified Heat Load Input Calculation SWP 811202004 was identified by DBVP as a calculation supporting ECN 5415.
The calculation was issued to determine reduced ERCW flow rates for ESF coolers when the ERCW is less than the maximum design temperature.
The calculation uses some heat loads that are less than the design or rated capacity of the coolers being evaluated. These input heat loads are described in the calculation as calculated room loads. However, these heat loads could not be verified by calculations, tests or other references.
The basis for these heat loads should be justified. _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ - _ _ _ _ _ _ _ _ _ _ -
TVA Response q
MEB has investigated the basis of the heat loads and found no reference for the input date.
A CAQR was written to address the resolution of this problem (SQNMEB8748, revision 1) and will be dispositioned in accordance with TVA's CAQ process.
Observation No. NEB ECCS Pump NPSH The calculation for the minimum water level in the containment sump, SQN-OSG7-008 entitled, Determination of Minimum Level in Containment Sump at Time of Switch Over to Recirculation Mode (B45-851218-235) resulted in a water level of 13.2 ft which is the top of the reactor vessel shield wall.
The calculation assumed that the following water inventories are introduced into the containment; RWST availabic volume, RCS, accumulators, and ice melt. This set of assumptions may not be limiting because they result from a large break 1
LOCA.
Since the purpose of the calculation is to determine minimum water level, the initiating event should be one which results in minimizing the water introduced into the containment.
In terms of minimizing water level, the limiting event might be a small break located high on the RCS which would reduce the ice melt contribution to water level. The calculation also assumed that the reactor cavity volume would not be filled during the event even though there are penetrations in the shield wall. The team feels that this assumption needs to be justified because it is non-conservative. The water level determined by this calculation along with the sump fluid temperature design value commitment is used in the evaluation of the NPSH available.
The team found in Section 6.3 of the FSAR that no credit was taken for subcooling of the sump fluid. The team also found that NUREG-00ll Supp. 1 (Sequoyah SER) gave relief from this commitment and established a sump fluid temperature of 1900F with no credit for containment pressure over atmospheric.
The SER (dated February 1980) concluded that the 1900F temperature would result in about 2.8 ft of H O excess NPSH availaile over 2
that required by the ECCS pumps. The FSAR should be updated to reflect the allowance for subcooling specified in the SER.
The team found that a fibrous insulation (NUKON) was added to the pressurizer power relief valve loop seal since OL.
Calculation NEB-841127-220 was performed to determine the effect of this insulation on the sump screen pressure drop if it were to become detached during a LOCA. The effect on pressure crop was determined to be 6.6 ft of H O when distributed according 2
to Reg. Guide 1.82.
This calculation post-dated the SER determination that resulted it. 2.8 ft of H O excess UPSH available assuming a 1900F sump 2
fluid temperature.
Included in this calculation was an analysis of the NPSH required 'or the ECCS pumps.
Additionally, this analysis did not use the potential reneut flow to detennine the NPSH required for the pumps. The potential runout flow is the intersection of the system resistance curve with the pump vendor certified head curve.
i Tha team feels th t ths NPSH calculational effort has not addressed the i
limiting case because of:
A.
the non-conservative assumptions in the water level calculation, B.
the potential impact of the fibrous insulation on sump screen pressure drop, and C.
the failure to consider pump runout flow to determine the NPSH required for the ECCS pumps.
TVA Response In response to this NRC observation, the existing calculation (SQN-0SG7-008) for the containment sump minimum water level has been revised to reflect that it applies specifically to a large Loss of Coolant Accident (LOCA). The calculation assumptions regarding location of sump water accumulation areas were revised to clarify that breaks that would result in filling of the reactor cavity were analyzed. A separate calculation (SQN-SQS4-0104, Revision 0) was issued to address a small LOCA that included assumptions similar to those in the large break calculation with the exception that no ice melt was credited for the limiting case.
The calculations assumed conservatively high Emergency Core Cooling System (ECCS) pump flows that clearly exceed the system design flows and are consistent with the conservatism identified in the SQN SER (NUREG-0011)
Supplement 1.
The resulting available ECCS net positive suction head (NPSH) calculation revision indicated adequate NPSH available.
0 Both calculations were based on a subcooled sump of 190 F consistent with the SQN SER Supplement 1.
The FSAR (section 6.3) will be revised in the 1988 amendment to be consistent with the SER regarding the method of NPSH determination.
The calculation that examined the effect of the fibrous insulation on sump flow and ECCS pump performance (SQN-SQS4-0084, Revision 0) is being revised.
Preliminary results for large breaks indicate that the additional flow reduction because of sump blockage is compensated by the fact that the sump 0
temperature may be assumed to be at 160 F since the break must occur in the vicinity of the pressurizer to dislodge the insulation.
A break in this location cannot directly interact with the sump, which is the basis for the 0
190 F assumption in the SER.
Small break NPSH and sump performance 0
calculations are based on a sump inlet temperature of 190 F.
Observation No. NEB Wide Range Containment Pressure Transmitters TVA calculation SQN-NAL4-002 Rev. 6 (B45870109235) stated that wide range containment pressure transmitters PT-30-310 and -311 have a range of -5 t u +60 psig with a 110.98 psi instrument error due to accident environment and seismic effects.
This calculation stated that these instruments were required by NUREG-0737 Section II.F.1, but that they were not required for mitigation of design basis events.
In addition, it was stated that these transmitters and their display indicators were not needed for operator action, and their t
use was not specified in the plant emergency procedures.
On these bases, the calculation determined that the instrument error of 110.98 psi was
)
acceptable. The team does not agree that the 116% error is appropriate since containment vacuum conditions could not be monitored.
In addition, use of these instruments should be prescribed in the plant emergency procedures.
The team is also concerned that the containment isolation provisions provided are not in compliance with the FSAR commitment (page 6.2-131) of double barriers for containment isolation.
TVA Response Pressure transmitters PDT-30-310 and -311 were added in accordance with the requirements of NUREG-0737. Their primary function is to provide indication of primary containment pressure above the design pressure, and they are required to have a span that is at least four times the maximum design pressure.
These transmitters are available to evaluate impending rupture of the containment in the event that pressure exceeds the design value. The transmitters have the stated accuracies.
This accuracy is considered to be acceptable in that additional safety-related containment pressure instrumentation is provided at SQN.
These safety-related instruments, PDT-30-44 and -45, cover the pressure range from -1 to 15 psig and are specifically addrecsed in the emergency response guidelines. The wide-range transmitters are therefore available for beyond design basis events for containment pressure trending purposes only.
This approach is consistent with the Westinghouse Owners Group Emergency Operating Procedures that have been approved by NRC.
As noted in S. A. White's letter to J. G. Keppler dated March 13, 1987, the containment is protected from vacuum conditions by vacuum breakers that are designed to open at -0.1 psig and are sized to limit the negative pressure within design conditions.
Nevertheless, as committed in the above letter from Mr. White to Mr. Koppler, TVA will explore replacing the instrument with a more accurate pressure transmitter. This activity is scheduled for resolution after unit 2 restart.
With regard to the FSAR containment isolation commitment for double barriers.
TVA has initia.ed PIR SQNEEB8795 to address this issue.
As part of the corrective action for this PIR, tests are being conducted to demonstrate that redundant pressure barriers exist between the containment environment and the annulus.
Observation No. NEB Essential Setpoint Calculations In HVAC calculation SQN-APS5-005 (B45860908236) there are five criteria provided for determining whether a setpoint calculation is essential or not.
One of the criteria used a redundancy argument in that the existence of a redundant instrument would, by irap11 cation, permit the instrument under review _ _ - _ _ _ _ _ _ _ _ - _ _ _ _ _ _ _ -
to be designated as non-safety related. The team does not agree that this is a valid criterion for such determinations.
During the team's review of this calculation, the redundancy criterion had not been used as a basis for classification of any HVAC instrument setpoint calculations.
TVA Response The subject HVAC calculation was revised on February 23, 1987, to delete the setpoint criteria in question. The inspection report notes that this criterion had not been used as a basis for classification of any HVAC instrument setpoint calculations. There are no other setpoint-related calculations performed in the HVAC section.
Observation No. EEB Battery and Charger Sizing The team is concerned about the adequacy of the installed system based on the team's review of calculation CPS-004 for class lE battery and charger sizing.
Specifically.
The evaluation of imposed loads on the batteries does not address motor in rush currents, worst case current requirements of 6.9 kv class 1E switch gear during load shedding, and random loads.
However, connected loads evaluated were assumed to be energized for the entire duration of the discharge.
The evaluation of the imposed load does not utilize the name plate rating of class 1E inverter that is rated for 20 kva, but uses only 15 kva that represents the attached loads.
The team found no control mechanism to limit the loading to 15 kva. The team noted that submergence calculation SQN-SBMG-1 uses the rated value of 20 kva and a trip setting of the inverter feeder breaker of 175 amps. At the minimum voltage of 105 v the inverter could draw an 18.5 kva load that is not consistent with 15 kva assumption used in calculation cps-004. This could result in overloading of the battery and charger, which would reduce the system's ability to supply the required voltage.
The charging time of 9 hours1.041667e-4 days <br />0.0025 hours <br />1.488095e-5 weeks <br />3.4245e-6 months <br /> was arrived at by using a charging efficiency of 75%.
This value was assumed and needs verification from the vendor, but was not listed as an unverified assumption.
The plant installed battery system has a lower capacity (14 positive plates) than the required capacity (16 positive plates) determined by recent TVA calculations. To justify the adequacy of the installed system, TVA has used a lower aging factor, making the end of life 89%
instead of 80% as originally calculated. However, the team noted that information regarding changing end of life (from 80% to 89%) was not passed to Sequoyah operations department when the calculation was completed in June 1986. This information was not passed to operations until another issue regarding battery performance was discovered during an INPO visit in November-December 1986. This demonstrates a breakdown in communication between the design organization and operations. c_________
The battery.is approximately 14' years.old; 20 years is considered to be the
~
normal life. Considering the age of the battery and also considering the omissions in the calculation'the team is concerned about the adequacy of the installed system.
Therefore, this calculation should be revised to include the following items:
Momentary in rush loads and random loads, a.
b.
Maximum rating of the connected loads.
In addition. TVA should verify that all loads added to the vital DC System as a result of ECNs are included in the battery sizing calculation. TVA should also verify by test prior to restart that calculated battery capacity can be supplied by the installed battery system.
TVA Response Inrush loading on the 125-volt de Vital Battery System consists primarily.
of switchgear trip and close coils, and de emergency lighting. While the i
switchgear loading may be determined analytically, little quantitative vendor data is available on de lighting inrush currents. Thus, subsequent to the subject NRC inspection in February 1987. TVA performed a special test to determine the inrush loading of the de emergency lighting cabinets powered from the 125-volt de vital batteries.
During the test, the normal loads for the de emergency lighting cabinets were also recorded, and in two cases were determined to exceed the values recorded in the circuit schedules on TVA battery board single-line design drawings.- These drawings were used as input to the development of the l
3 battery duty cycle in DNE calculation SQN-CPS-004 Thus, the conservatism and adequacy of the duty cycle evaluated in SQN-CPS-004 became indeterminate. This deficiency in the single-line circuit schedules has been identified as a CAQ and appropriately documented in CAQR SQT870791, in accordance with the TVA CAQ process. As a portion of the corrective action for this deficiency, TVA has revised calculation SQN-CPS-004 to include a detailed load analysis of the most heavily loaded 125-volt de vital battery.
This revision includes a duty cycle based upon the worst-case composite operational mode for the battery and includes the inrush requirements as determined analytically and 4
empirically.
The results of the calculation verify sufficient battery capacity for SQN unit 2 restart with no operational constraints, and no technical specification revisions are necessary.
The single-line drawings' circuit schedules will be revised for Engineering Change Notice (ECN) L7186 to correct this documentational deficiency. _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ - _ _ - _
While th3 Sequoyah vital invarters are rated to deliver 20 kVA, thair maximum design load is limited to 15 kVA (125 A at 120-V ac).
This limitation is clearly documented in FSAR section 8.3.1.2.2 under subparts entitled "Uninterruptible Power Supply" and "120 V AC Vital Instrument Power Distribution Boards." It is also clearly documented in DNE calculation SQN-CPS-004 Revision 2.
To provido additional documentation, however, the vital de control power design criteria, SQN-DC-V-11.6, has been revised by design input memorandum (DIM) SQN-DC-V-ll.6-5 to reflect the 15-kVA maximum design load limit.
Further, TVA electrical drawings 45N703-1,
-2,
-3, and -4, and 4SN706-1,
-2,
-3, and -4 have been revised to note the loading limit. This load limit is strictly controlled in accordance with SQEP-09, " Change Review Checklist for Electrical calculations."
The SQN vital inverters have an operational officiency of 80 percent.
Further, the power factor of the inverter load is approximately 0.80:
DNE calculation SQN-VP-VAC-2 documents a measured value of 0.768.
Given the documented load limit of 15 kVA (which corresponds to a current of 125 A at 120 V), the battery output current, IB, may be determined as follows:
(Pout}INV = (Pout) BAT (EFF)
(120 V)(125 A)(0.80) = (105 V)(I )(0.80)
B I3 = 142.86 A This current is less than the 150 A used in the determination of vital battery capacity; thus, the vital battery calculation adequately addresses inverter loading.
The vital battery recharging time was calculated in accordance with TVA Electrical Design Standard DS-E3.1.1, which uses an assumed value for battery recharging efficiency.
The calculation has been revised and issued to use industry standard data instead.
i Revised DNE calculation SQN-CPS-004 Revision 2, in conjunction with recent vital battery service tests, verifies adequate battery capacity for unit 2 restart without capacity-related operational constraints.
Observation No. EEB Breaker Coordination The team's review of calculation APS-003 for the breaker coordination study indicated that TVA found that the feeder breakers for the ERCW 480V boards 1A-A, IB-B, 2A-A, 2B-B, and the feeder breakers for the diesel generator 480V boards lAl-A, IA2-A, IB1-B, 1B2-B, 2Al-A, 2A2-A, 2B1-B and 2B2-B are not coordinated with the individual load breakers. The team reviewed TVA's corrective actions and noted that TVA intends to reset the feeder breakers.
Review of the breaker coordination curves for diesel generator 480V boards.
revealed that resetting the breakers will not resolve the problem since the breakers cannot be reset to the appropriate valves.
It would appear that these breakers either have to be replaced or the sensor mechanism has to be replaced. TVA intends to perform this corrective action after restart due to the fact that the loss of one board in ERCW and DG system would not degrade the safe shutdown capability of the plant. The team found that the calculated instantaneous trip settings (700%) do not agree with the trip settings shown on the drawings (900%) #3591 A12 Rev. 5, 3591A 14 Rev. 4, 3591A 16 Rev. 6, 3591A 18 Rev. 4, 3591 A20 Rev. 4, 3591 A22 Rev.
4, 3591A 24 Rev. 6 and 3591A 27 Rev. 3.
TVA informed the team that this mistake was due to improper verification of installed breaker data by the walkdown team.
TVA Response i
The normal design procedure requires the preparation of an ECN to resolve a deficiency identified by a calculation.
During the design process (which involves preparation, verification, review, and approval) the appropriate corrective action is determined and is incorporated into the design output drawings in accordance with the issued ECN.
Further, in support of the ECN, the calculation that identified the deficiency is revised in accordance with the design output drawings to incorporate the corrective action and to verify its adequacy.
Thus, the actual corrective action is determined not by the calculation, but by following the ECN procedure.
Ultima tely, by following this standard procedure, the observation noted by NRC would have been detected and corrected by TVA; thus, no specific corrective action is required for this observation.
Observation No. ERB 120V AC and DC Solenoid Valve Voltage The team's review of calculation SQN-cps-001 for 120V AC and 125V DC solenoid valve voltage study indicated the following:
The calculation of cable impedance does not address the effects of junction boxes, electric conduit seal assemblies, cable slacks and higher temperature (ambient) of operation during accident.
In the 120V AC voltage study, assumptions were not listed in a dedicated section but were scattered all over in the calculation and were indicated by a word " assume" in parenthesis next to the value assumed.
EEB personnel indicated to the team that all such unverified assumptions will be verified before restart.
The team believes if all assumptions were listed in a dedicated section, the chance of some assumptions not being verified (due to being buried in the calculation) will be reduced.
TVA Response The calculation has been revised to address harsh environments, conduit seals, and junction boxes...
l TVA EA audit 86-23 identified documentational departures from the requirements j
of NEP-3.1 in several electrical calculations.
The auditor stated, however, j
that not one of these departures could be considered of major significance.
j Since the technical adequacy of the calculations has been independently verified by TVA during the review and issue procedure and by Sargent and Lundy during the electrical calculations program assessment, no effort is planned to j
generally revise the existing documentation.
During the ongoing effort to j
verify all previously " unverified assumptions," all assumptions will be reviewed and the calculations will be revised to locate the assumptions in a single dedicated section.
This effort will be complete before SQN unit 2 restart. Further, to ensure future documentational adequacy, all calculations personnel received training in preparing, checking, and reviewing calculations in accordance with TVA's Nuclear Engineering Procedures.
This training was completed by February 20, 1987.
Observation No. EEB Setpoint Accuracy Calculation for Replacement of Rosemount with Gould Transmitters The setpoint accuracy calculation for the containment annulus differential pressure transmitters PDT-65-80, -82, -90, and -97 was provided in I&C Calculation File 29 (RIMS B43850830903). This calculation addressed Rosemount transmitters.
An ECN had been initiated to replace these transmitters with Gould transmitters for environmental qualification purposes.
The setpoint accuracy calculation was not updated to reflect use of Gould transmitters.
TVA Response The calculation for the existing Rosemount transmitter was revised and issued on February 5,1987, to state that the calculation should be superseded by the Gould calculation when the changeout occurs.
The Gould calculation already contained this statement.
Observation No. EEB Assumed Value Error for Sensor Measurement and Test Equipment Accuracy In the setpoint accuracy calculation for RWST level transmitters LT-63-50,
-51, -52, and -53 (RIMS B43860228901), the assumed value for sensor measurement and test equipment accuracy (SMTE) was listed as 10.05 percent of span rather than 10.5 percent of span.
In the numeric portion of the calculation, the SMTE term actually used the correct value of 0.5 percent.
Hence, this appears to be a single random human error that had no impact on the calculation results.
TVA Reeponse The calculation was revised and reissued on February 9, 1987. _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ __
CEB OBSERVATIONS Observation No. CEB Rigorous piping Analysis N2-67-8A I
The geometry for this piping analysis consists of an L-shaped branch line composed of 1-inch and 2-inch diameter pipe with a 1-inch by 2-inch relief valve located at the branch line elbow. This branch line is supporte4 from 24-inch ERCW run pipe located in the auxiliary building.
The team noted the following:
(1) The piping dimensions and configuration detailed on piping stress isometric 47K450-115 Rev. 2 are not completely defined on piping physical drawing 47W450-4 Rev. 45.
Only the larger 10-foot 3-inch span dimension for the 1-inch diameter line is detailed, and an elevation for this span is not shown.
In addition, a designation for the valve is not provided.
(2) The piping stress isometric details a span length of 8-feet 9-1/4-inches for the 1-inch diameter pipe, which is less than the 10-foot 3-inch span length detailed on the piping physical drawing.
(3) The 1-inch by 2-inch TVA Class C relief valve was purchased under the requirements of TVA procurement specification 72C53-92795-10.
Attachments A and B of the specification exempt TVA Class B and Class C valves less than 6-inches in diameter from the seismic qualification criteria which are stipulated for the valve vendor in Appendix F, Guide for Qualification of Seismic Class I and Seismic Class II Mechanical and Electrical Equipment, which forms a part of the procurement document.
This exemption is not consistent with the rigidity requirements specified for TVA Class B and Class C valves in FSAR Tables 3.9.2-1 and
-3, or FSAR Table 3.2.1-2, which indicates that TVA Class C valves in the ERCW system have been seismically qualified by test, or SQB-DC-V-7.4, ERCW Design Criteria, Section 3.7.1.2, Seismic Requirement, which requires the ERCW system to be designed to seismic Category I criteria.
TVA Response to (1) and (2)
TVA agrees with the observation findings. A field walkdown was performed during February 1987, and the "as-built" dimensions have been incorporated into a study run.
Results of the study run show that the piping remains qualified to all design allowables.
CAQR SQp870094 has been written to document, track, and correct the observation findings and is discussed below.
Root Cause - All piping, two inches and smaller, on the Essential Raw Cooling Water (ERCW) system was originally field routed.
It appears that the Office of Construction could not easily support this line in accordance with the Alternate Analysis Criteria and requested that it be rigorously analyzed.
Since the line was field routed, the physical piping drawings provided few dimensional details. The physical drawings should have been revised to show all required dimensions.
It is unknown how the analyst obtained the missing dimensions.
Since it was normal practice to show all required dimensions on the physicals for rigorously analyzed piping, this is assumed to be an individual error. __________.m___
Tha analysis was in agreement with the original dimensions that were shown on the physical drawings. The physical drawings were later revised to show the 10-foot 3-inch dimension. No documentation exists to show that the new dimension was evaluated for its effect on the analysis, and the analysis isometric drawing was not revised to show the correct dimension.
i Recurrence Control Actions - SQEP-13 has been issued and requires the "as-built" and "as-designed" drawings be compared and resolved into one drawing change authorization (DCA) that, upon completion of construction, will be issued as a Configuration Control Drawing. All analysis drawings will be in agreement with the Configuration Control Drawing upon issue. This procedure will control all new piping and piping modifications (reroutes) and should prevent recurrence of the problems cited.
Corrective Actions 1.
Revise the physical pipin6 drawings to adequately dimension the subject piping.
l 2.
Revise the analysis documentation to identify the valve and address and document the dimensional discrepancies.
l TVA Response to (3)
All safety-related valves at SQN are required to be seismically qualified, but not necessarily by the valve vendor.
TVA has taken the responsibility for seismic qualification of the handwheel-operated manual globe valves of four-inch and less nomiral diameter.
The valve vendors are not required to fulfill any special seismic requirements for such valves.
This practice was initiated on the basis of Westinghouse studies that are documented in their topical report WCAP-7700R1 entitled, " Seismic Analysis of Nuclear Power Plant Auxiliary Equipment." The American Society of Mechanical Engineers (ASME)
Code,Section III, paragraph NB3513, provided the basis for the four-inch and smaller size limitation. ASME Code Case N-62 Figure 2 - Globe Valves, depicts the types of valves for which this approach has been applied.
Such valves possess inherent ruggedness sufficient to withstand device seismic qualification levels, 3g(H) and 2g(V), and they are inherently rigid (f' greater than 25Hz). A Civil-Engineering Branch (CEB) Report or calculation will be issued to fully document this justification by July 15, 1987.
For the one-inch by two-inch relief valve in question under Observation CEB-1(3), the exemption of seismic qualification requirements is a misapplication of the approach described above. Accordingly, this situation presents a potentially generic condition whereby other procurement of small valves (less than six inches nominal pipe size) could have been made without appropriate seismic documentation. As a result, CEB has originated CAQR SQS870070 to document this deficiency and define corrective action relative to the total generic scope.
Specifically, corrective action for this CAQ will involve:
l' Verification and documentation of seismic qualification for the one-inch by two-inch relief valve (by CEB).
2.
Review of other valve procurement specifications that could have resulted in similar deficiencies at SQN (by EEB and MEB)..
3.
Verification and documentation of seismic qualification for all cases identified by the review noted in item 2 (by CEB).
New seismic qualification criteria for SQN valves and other in-line fluid system components will be issued af ter unit 2 restart, in conjunction with the SQN DBVP. These criteria will be patterned after WB-DC-40-31.12 and will contain appropriate clarification regarding this issue.
As stated in previous correspondence to NRC, the design criteria included in the postrestart phase of the DBVP are scheduled to be issued by December 31, 1987.
NOTE:
The following five observations relate to miscellaneous steel calculation package, " Auxiliary Building Access and Platforms (PMP 840625 808); drawings 48N1210 through 48N1216." The observation's are addressed in one response following observation CEB 6.
Observation No. CEB Structural Steel Sizing Calculations The review of TVA calculation PWP 840625 808 auxiliary building access and support platforms and stairs, showed that various steel members shown on drawings 48N1210, 48N1211, 48N1213 and 4BN1214 do not have any calculations.
There is no design documentation supporting the structural adequacy of these members.
Observation No. CEB Structural Steel Details During the review of TVA calculation PWP 840625 808 auxiliary building access and support platforms and stairs, the team noted that there was a lack of evaluation for structural steel connections and details.
TVA uses welded and bolted connections for the structural steel in the auxiliary building.
The calculations reviewed for 48N1210, 48N1211, 48N1213, 48N1214, 48N1215, 48N1216, and 48N1216-1 did not show evaluations for welds and bolts. Also, the stress calculations for the stiffener plates used in drawing 48N1216 Detail A could not be located, i
Observation No. CEB Platform Steel Calculations and Drawings The review of TVA calculation PWP 840625 808, addressing auxiliary building access and support platforms and stairs, showed that for drawing 48N1214 one of the structural members was calculated to be a channel section 9[13.4.
The review of drawing 48N1214 Rev. 3, Plan at elevation 692'-6", shows this member to be a channel section 8[11.5.
Therefore, a member smaller than what was required by calculations was installed.
A similar situation exists for computer models prepared for the analysis of steel platform at elevation 724'-3".
The calculations performed for drawing 48N1216-1 shows that the two computer models prepared do not match the configuration of the platform as shown on drawing 48N1216-1 Rev. 11.
l Observation No. CEB Revisions to Steel Platform Calculations The review of TVA calculation PRP 840625 808, addressing auxiliary building access and support platforms and stairs, showed that calculations related to drawing APN1215 were revised to add additional hanger loads. The structural steel was analyzed for adequacy to carry the new loads, however, this reanalysis was not checked or verified. Also, the components which resulted in the Sdditional support loads were not identified.
Observation No. CEB Seismic Loads for Steel Platforcas The review of TVA calculation PWP 840625 808, addressing auxiliary building access and support platforms and stairs, showed that steel platforms were not consistently designed for seismic loads.
In certain cases, seismic loads were not included in the load combination to determine structural adequacy. This can be seen in the calculations performed fo'r drawings 48N1211, 48N1213 and 48N1214.
In addition, calculations for drawing 48N1216 showed that steel stress allowables were lowered by 1/3 to account for seismic loads. A later revision used the full value of the allowable to evaluate additional conduit loads without consideration of seismic loads.
No justification was provided either for lowering the stress allowables or for subsequently raising them.
The rame calculation assumed the platforms were rigid to determine seismic loads. There were no substantiating frequency calculations to justify the rigidity assumptions.
TVA Response An existing CAQ (SCR SQNCEB8711, Revision 1) addresses the above and the following related issues:
Some platform calculations do not include qualification for field-routed attachments. Also, some platform calculations do not include documentation to show that the latest revisions for pipe supports for rigorously analyzed piping have been considered.
Some field changes that affect the configuration of the platform have not been incorporated into the design calculations and drawings.
This condition described above is also applicable to miscellaneous and structural steel.
In general, a review of the estimated and assumed loading in the construction / design process as specified in Design Criteria SQN-DC-V-1.3.3.1 has not been performed for most miscellaneous snd structural steel.
SCR SQNCEB8711, Revision 1, describes corrective actions, some of which have been completed, as follows:
Platforms In order to determine the effect that the described deficiencies have on the qualification of the platforms, a reanalysis is being performed to be completed before restart for a minimum of five platforms. o
Tha fiva platforms that are being reanalyzed have been selected to be representative of the in-place conditions. The selection process has been documented.
Of the five representative platforms, two are the m)st heavily loaded platforms that were recently included in the thermal evaluation (SCR SQNCEB86103) from the Reactor Building, and three platforms were selected from the Auxiliary Building.
A walkdown procedure has been developed, issued, and implemented to obtain "as-constructed" information concerning the identified deficiencies.
The procedure requires visual verification that the weld configuration, location, and length are compatible with the drawing requirements.
The platforms are being analyzed considering the "as-built" information to determine the adequacy of the platforms in meeting the design criteria requirements.
If all of the analyzed platforms meet the design criteria requirements, then the platforms will be acceptable for restart.
If any of the analyzed platforms do not meet the design criteria requirements, the platforms will be evaluated in accordance with the Restart Requirement Criteria, Based upon the number and type of deviations, the neld to evaluate additional platforms will be made.
Deviations will also be addressed through follow-up responses. Any platforms not meeting the Restart Requirement Criteria will be modified before restart to meet the design criteria requirements.
Status:
Analysis of the as-built configuration has been performed and checked for three of the five platforms, and preliminary analysis has been performed for the remaining two.
To date, a connection plate and an anchor bolt for a surface-mounted plate do not meet restart criteria and are undergoing final review. The other facets of this platform and the other two platforms meet restart criteria requirements.
Miscellaneous Steel (Excluding Platforms)
The adequacy of the design of the remaining miscellaneous steel is being verified through the technical review of 55 randomly selected features
(
Reference:
Detailed Technical Review Plan SQN-CEB-87-02.
As part of the resolution of SCR SQNCEB8711 Revision 1, five randomly selected structures were examined to determine if there are significant attachments that were not considered in the design calculations or if there were any significant changes in configuration.
Status: The review of the five features is complete, and it has been determined that there were no significant attachments that have not been considered in the design and no significant changes in configuration.
Based on this evaluation, the miscellaneous steel (excluding platforms) is determined to be acceptable for restart for significant attachment loadings and configuration changes. The Technical Review Plan identified five CAQRs i
that require evaluation before restart. Postrostart work identified in the Detailed Technical Review Plan SQNCEB87-02 will be completed as defined by closure of the Detailed Technical Review Plan.
) _ - _ - _ _ _ _ _ _ - - _ - _ _ -
Structural Steel Because structural steel is generally of heavier type construction, it can be generalized that it is not as sensitive to additional attachments. Also, there are generally detailed fabrication and erection drawings that lessen the chance for field changes. Therefore, an examination of portions of five structural steel features will be performed before restart. The purpose of this examination will be to determine if attachments that were not enveloped in the original design have been made to the structure.
The examination will also determine if the configuration is significantly different from the drawing.
If it is determined that there are attachments that were not enveloped in the original design or if significant configuration differences exist, then the structure will be evaluated considering these aspects.
Status:
Based on the evaluations to date, the structural steel features have been determined to be acceptable for attachments and configuration changes.
Actions Taken to Prevent Recurrence The identified deficiencies involve inadequate configuration control (i.e.,
control of attachments to miscellaneous and structural steel) and inadequate or incomplete design calculations.
Control of attachments to miscellaneous and structural steel is now addressed by the following:
1.
NEP 6.1, for change control.
2.
SQN instruction, M&AI-11, for the control of attachments to miscellaneous and structural steel.
3.
SQEP Al-llA, for field change requests (FCRs).
4.
SQEP-13, Transitional Desig'n Change Control.
5.
SQEP-66, Attachments to Civil Features.
The adequacy of civil discipline calculations is addressed by Policy Memorandum (pM) 86-02.
Observation No. CEB Rigorous Piping Analysis N2-67-3A-4 The piping geometry for this analysis primarily consists of 6-inch diameter ERCW pipe which is supported from the containment building at elevation 682-feet 6-inches and is rigidly attached to the steel containment vessel (SCV) at penetrations X-59 and X-63.
The team found that:
(1) Rev. O of the piping analysis documented acceptance of accelerations greater than 2 and 3 g for valve 2-FCV-67-88 adjacent to penetration X-59.
Rev. 1 of the piping analysis evaluated a valve operator change due to ECN L-5824 (replacement of motor operators due to lack of NUREG-0588 environmental qualification of motor insulation). TVA performed a " study" to evaluate the effects of increased valve operator weight and center of gravity dimension or, the valve dynamic response, but did not document this evaluation in Rev. 1 of the piping analysis.
(2)
Rev.'2 of the piping analysis references four CEB "as-built" pipe. support gap evaluations in accordance with ECN L-6706.
The team found errors in two of the four "as-built" evaluations. The calculation to determine if the 1-15/16-inch "as-built" radial gap for pipe support HERCW-10 could accommodate the pipe movement at that location we.s incorrectly performed. No calculation was performed to evaluate the 3/4-inch "as-built" radial gap for pipe support HERCW-14.
(3) The team checked the documents which qualify pipe penetrations X-59 and X-63 and found that the original qualification documents for the Unit 2 penetrations had been incorrectly superseded by the qualification documents which CEB prepared for the corresponding Unit 1 penetrations.
TVA Response (1) TVA agrees with the observation finding except as follows: The " study" was documented in the analysis calculation package; however, the computer analysis printout and the revised valve accelerations were not documented.
The analysis computer study run was recreated and the valve accelerations were shown to be less than the accelerations previously approved.
CAQR SQP870180, Revision 1, has been written to document, l
track, and correct the observation finding and is discussed below.
1 Root Cause - No procedural requirement existed that required that detailed qualification results for evaluations (i.e., computer analysis printout, component qualification review) be documented in the analysis calculation package.
Proposed Recurrence Control Action - The SQN Rigorous Analysis Handbook will be revised to require that detailed qualification results be documented when performing analysis evaluations.
proposed Corrective Action - Revise the analysis calculation packago to document the detailed qualification results.
(2) TVA agrees with the observation findings. A more detailed computer evaluation for the gap at support No. HERCW-10 shows that the existing clearance is adequate.
CAQR SQp870096 has been written to document, track, and correct the finding.
l L
Th2 inndequate e,cp noted at support No. HERCW-14 was identified from a discrepancy in ite SMI 1-317-24 inspection program.
An evaluation of the gap at support No. HERCW-14 has shown that insufficient clearance exists to accommodate the pipe movement in the unrestrained direction.
PIR SQNCEB8730 has been written to document and track this finding.
Root Cause - The analyst incorrectly calculated the pipe movement at support No. HERCW-10 in the analysis calculation package.
The support No. HERCW-14 was designed with inadequate gap in the unrestrained direction in accordance with the issued support load table.
The design error is believed to be an individual mistake. The support was fabricated with less gap than specified by the support detailed design drawing.
Recurrence Control Action - Adequate procedures were and are in place to ensure that the support design accommodates the pipe movement in the unsupported directions.
Corrective Action - The analysis calculation N2-67-3A4 has been revised to correct the error for support No. HERCW-10.
The support design calculation and drawing for support No. HERCW-14 and the field modification will be completed as part of the SMI 1-317-24 program (ECN L5706).
Other support gap inadequacies identified as part of the SMI 1-317-24 program and other inspections are being documented by separate CAQRs such as SQP870441.
(3) During the inspection of SQN calculations, the NRC team selected the ERCW system on unit 2 to review in detail. In the course of this review, it was discovered that the penetration loads in the piping calculations packages did not match the loads that qualified mechanical penetrations X-59 and X-63.
Upon investigation. TVA determined that a miscommunication had occurred in the transmission of penetration loads.
Multiple requests (some applying to single units and some to both units 1 and 2) were attached to the transmittal. The transmittal cover was labeled units 1 and 2; however, the load sheet for the penetration cited was labeled as unit 1 only. The evaluation mistakenly applied the loads to both units, pIR SQNCEB8734 has been issued documenting this condition. As part of the corrective action, a complete list of qualified mechanical penetrations has been compiled.
This list provides the problem numbers and checked dates for the Auxiliary Building and Reactor Building l
problems for each penetration in which loads were sont for requalification. This data was forwarded to the SQEP piping analysis section for their review and verification against what they have in their calculations packages. Also included in the data were hard copics of the penetration load summaries. _ _ - _ __ _
The 3QEP piping analysis section reviewed the data and provided a marked-up copy of the list defining all of the discrepancies on each unit. Discrepancies have been identified, and the affected penetration calculation packages are being revised to reflect the latest loads.
This effort will serve as the baseline for penetrations currently qualified.
Observation No. CEB Qualification of Seismic Category I Buried Pipe The team asked CEB to provide the criteria used to qualify the seismic Category I buried ERCW pipe which runs between the ERCW pumphouse and the auxiliary building, and the associated calculations.
CEB indicated that buried pipe is seismically qualified in accordance with design criteria SQN-DC-V-13.5, Design Criteria for Seismically Qualifying Buried Piping Systems, which CEB issued on September 5, 1972.
However, CEB cannot retrieve the calculations which document the seismic qualification of the buried ERCW pipe. Moreover, CEB did not identify these
" essential" calculations as missing to evaluate and assess the need for regeneration of these calculations before restart of SQN Unit 2.
TVA Response During the inspection of SQN calculations, the NRC team requested calculations relating to the seismic analysis of buried ERCW piping as part of a detailed review of the unit 2 ERCW system. An extensive search was conducted for the calculations.
Calculations were not retrievable.
The missing calculations are considered to fall within the scope of SCR SQNCEB8714.
Missing calculations are regenerated using methods described in the applicable design criteria.
For buried piping, the appropriate design criteria are SQN-DC-V-13.5, " Seismically Qualifying Puried Piping Systems." The design criterla require that calculations be made to:
a.
check stresses induced by seismic loading to ensure that ASME Code allowLbles are not exceeded, and b.
ensure that differential movements and resulting stresses are properly accommodated at building penetrations and at interface between fill and natural soil.
)
Calculations for the ERCW piping have been made for increased stress resulting from seismic loads.
All penetrations and interfaces have been identified and have been evaluated.
l Seismic analysis of the buried fire protection system piping has also been i
made.
i.__-_____-____-____A
All analyses related to this observation have been completed and documented.
The analysis results indicate full compliance with the design criteria.
Procedures are in place to prevent the described condition from recurring.
NEP-3.1 requires that calculations be performed on all design input or output documents needing development, proof, or support.
NEP-3.1 also requires the lead engineer or group head to determine the need for calculations.
CEB PM 86-02 gives directives to CEB personnel in calculations preparation and handling. Training in the use of procedures is required by NEP-1.2.
NEP-3.1 requires the group head or lead engineer to ensure that calculations are issued in accordance with NEP-1.3 (records control).
Observation No. CEB Reinforcing Bar Cut Evaluation The review of TVA calculation PWP 840920 705, south main steam valve room auxiliary building, showed that two reinforcing bars were cut and documented on FCR 2255. An evaluation of the slab was made for these cuts to determine structural adequacy. This evaluation failed to consider the seismic load on the slab.
During the inspection, TVA performed additional calculations to show that the slab is structurally adequate to carry all the loads including seismic.
TyA Response l
This observation identified a deficiency in a 1984 revision to the calculation l
package PWP 840920 705, page 80B, for the 706.0 slab (ori;iaal calculations dated 1971) to evaluate requested reinforcing bar cuts.
The revision did not consider earthquake loadings.
A PIR has been issued (SQNCEB8754) that identifies the following corrective i
l actions:
l l
A.
Evaluate the effects of cutting the rebars included in FCR 2782 when earthquake loads are included.
B.
Revise calculation book "A6" and all other original calculation books to cross-reference new calculations recently completed in accordance with SQN-DC-V-1.3.3.1 that evaluated the cumulative effects of rebar cuts.
Also, include the calculation for the evaluation made in (A) above, in this revision of calculation book "A6."
NOTE:
As a result of the overall evaluation of the cumulative effects of rebar cuts, which did consider seismic loadings, it is not necessary to review all the original calculations for the condition. This evaluation was performed as a result of the approved Corrective Action Plan (CAP) for responding to Employee Concern Task Group Element Report 215.2. _ _ _ _ _ _ _ _ _ - _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ - _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ - _ -
Bsekground Information:
1.
The original calculations were completed in 1971 using ACI 318-63 working stress design methods. At that time, the allowable stresses for reinforcing steel were 0.4fy (D+L), 0.5fy (D+L+0BE), and 0.9fy (D+L+2*0BE) as set forth in design criteria SQN-DC-V-1.1.
Using the increased allowables, the controlling load combination was D+L.
2.
The calculations were revised in 1984 for evaluation of specific rebar cuts for FCR 2782. The designer reviewed the original design for the slab (as the new calculation was placed directly behind the calculation sheets for the original design of this' slab) and noted that the original design was for D+L only, not realizing the reasoning behind its use.
The evaluation for the rebar cuts used Ultimate Design method as specified in design criteria SQN-DC-V-1.3.3.1 for which no increase in the allowable reinforcing steel stress for the D+L&OBE load combination is permitted.
Therefore, OBE should have been considered in evaluating this FCR.
3.
To evaluate this error, a single-page calculation using Ultimate Design method considering OBE was. developed. The results indicated a very minor increase in moments (four percent).
This did not affect the structural q
integrity of the slab.
The remainder of the calculation book was reviewed, and this error had not
{
been repeated, even though other revisions had been made for other rebar cuts. Based upon the fact that this calculation was an independent revision (i.e., made for a particular FCR) and that the error had not been repeated in the calculation book, TVA considers it to be an isolated occurrence.
Observation No. CEB Weld Evaluation for Conduit Support TVA calculation B25 850304 300 was performed to design conduit support MK115 for FCR 2420.
A detailed computer analysis for the welded structural frame showed the steel member stresses were within allowables.
Since all the structural members were connected by welds, an evaluation of the welds for maximum moment and shear should have been performed. The team could not locate such weld evaluations.
TVA Response No weld calculations were included for MK 115 in calculation package B25 850304 300.
Calculation B25 870113 300, sheets 13 of 46 and 16 of 46, show typical weld design for drawings 48N1313-1 through 10 that includes 48N1313-9 for MK 115. These calculations should have been referenced in the calculations for MK 115.
Discussion with the designer indicated that he believed that, since the welds fully developed the members and the member stresses were so low, engineering judgement would qualify the welds. However, this was not documented in the calculation. i
In order to resolve the issus, wald calculations were performed and checked as a revision to the calculations. These calculations are conservative and show the all-around welds to be adoquate.
The welds for all the other connections in this drawing series are covered by the sheets 13 and 16 of 46 referred to earlier.
A PIR was initiated on this issue (PIR SQNCEB8755). The corrective action for this PIR has been completed as described above.
A DNE Interim Order has been issued as a supplement to NEP 3.1 to require engineering judgement to be documented. This error was minor and could have been avoided if the designer had either referenced the calculations for the typical welds or stated in the calculations his engineering judgement.
Observation No. CEB Pipe Rupture Evaluation for Concrete TVA calculation PWP 840920 705, south main steam valve room auxiliary building, was revised in 1976'to accommodate the changes in pipe rupture loads. An evaluation of the floor slab was made by assuming that the concrete compressive strength would be 6500 psi compared to the original design value of 4000 psi. A justification for this assumption was not documented.
Even with assuming high strength for concrete, the calculations showed that in certain areas the concrete and steel allowable stresses were exceeded by 19 and 16%, respectively.
Stresses above the allowables were accepted by both the calculation preparer and checker without any documented technical justification.
TVA Response A review of concrete slab calculations for the west steam valve room revealed deficiencies related to unverified assumptions for concrete strength and reduction in pressure loads.
The designer assumed a concrete strength greater than called for on the concrete outline drawings and assumed that design pressures from pipe break would later be reduced. These assumptions were used to justify an overstress condition for the concrete shear and rebre calculations.
A CAQR (SQP870183) has been written that identifies the following recommended corrective actions:
1.
Determine the in-place 90-day equivalent concrete strengths for the concrete pour for the unit 1 and unit 2 west steam valve rooms.
2.
Establish final pressure loadings. l
3.
Review othsr category 1 or safety-related structural calculations outside the Reactor Building and east steam valve room for calculations using extreme pressure loads and/or pipe rupture loads where designers may have used increased concrete strengths or unverified assumptions regarding design accident pressures. Review of concrete design calculations for the concrete quality evaluation in the Reactor Building and east steam valve room did not identify other cases where unverified assumptions may have been used to justify overstress conditions.
The evaluation of those structures concluded that the structures are adequate to perform their intended functions. This review is documented in Exhibit F of Report CEB 86-19-C.
Review of 47E235-series environmental drawings will identify areas of extreme pressure loads outside the Reactor Building and east steam valve room.
Actions 1 and 2 above will be utilized to verify or substantiate assumptions in the original calculations. Action 3 will be utilized to define the extent of unverified assumption usage to dismiss overstress conditions.
Action to Prevent Recurrence 1.
The recently completed SQN concrete quality evaluation documented in Exhibit F of CEB report 86-19-C documents the requirements for use of increased concrete strength based on estimated in-place concrete strengths derived from actual cylinder tests.
Design Criteria and Design Standard revisions have been issued that will give necessary direction to designers.
2.
Engineering procedures NEP 3.1 and 3.2 have been issued that provide direction on properly documenting the use of engineering judgement.
3..
The engineering procedure on preparation of calculations, NEP 3.1, was revised to require documentation on the calculation cover sheet if an unverified assumption is used in a design calculation.
Unverified assumptions are tracked, in accordance with this procedure, so that they will be verified.
Current Status of Corrective Actions The following items are numbered the same as the corrective actions for CAQR SQP870183 listed above:
1.
From the concrete pour records it was determined that the actual in-place 90-day equivalent strengths are 6,100 psi and 6,300 psi for unit I and unit 2, respectively.
2.
A reanalysis of the design pressures in the west steam valve room was performed.
The results of this reanalysis are documented in a quality information release (QIR).
Using the results of items 1 and 2 above, the calculations for the elevation 706 floor slab for the west steam valve room have been revised and checked.
Design stresses are within allowables. _ _ _ _ _ - _ - _ _ _ - _ _ _ _ - _ _ _
3.
Correctiva action for this item is underway and will be completed in accordance with requirements of the CAQ process.
Generic Evaluation Regarding possibility of Similar Discrepancies in Other SON Calculations As noted in item 3 of the corrective action for CAQR SQp870183, the Reactor Building and east steam valve room concrete calculations were recently reviewed as part of the concreto quality evaluation, and any unverified assumptions regarding reductions in accident pressure loads or assumed increased concrete strength with age would have been picked up in this review. The calculations contain most of the areas where increased concrete strength was needed in the concrete design in order to qualify structures for extreme accident pressure loads. Documentation of this review is summarized in Exhibit F of report CEB 86-19-C that was transmitted to NRC on February 6,1987 and the following calculation packages that summarize the results of this review:
Summary of Structural Evaluations of Category I Structures for In-place Concrete Strengths and Review of Concrete Anchorages and Embedded plates for Concrete Strengths Less Than Original Design Requirements. The only areas that were not reviewed were a limited number of areas in the Reactor Building with 3000- and 4000-psi concrete.
The reason the west steam valve room was not reviewed as part of the concrete quality evaluation was that the actual in-place strengths were greater than the strengths specified on the design drawings. Also, it is believed that this is the only area outside of the Reactor Building and east steam valve room where extreme accident pressures occur; however, as identified in the corrective action, the 47E235-series environmental drawings are undergoing review to confirm this.
Since the NRC inspection on February 2-13, 1987, in Knoxville, the Reactor Building, east steam valve room, and Auxiliary Building calculations were reviewed for calculations performed by the same designer that made the unverified assumptions in the west steam valve room calculations.
The results of this revealed other calculations performed by this designer; however, no unverified assumptions were found in this further review.
Observation No. CEB Use of Variable Damping for Conduits TVA design criteria SQN-DC-V-13.10, seismically qualifying conduit supports, was revised on 11/20/85 to include the span lengths and the support loads as i
developed in TVA calculation B41 851105 028.
A review of this calculation l
showed that a variable damping ratio was used in determining the seismic loads l
on the conduit supports. TVA used a damping value of 2% for frequencies I
greater than 10 Hz and 5% for frequencies less than 10 Hz.
TVA's commitment, as shown on Table 3.7.2.4 of the Sequoyah FSAR, is a constant 2% damping value for the safe shutdovm earthquake. The use of a higher damping value would lower the conduit support loads and might yield an unconservative design. _________ _
TVA Ruponsw The fourth sentence in the observation is not correct:
1.
The FSAR reference is for piping, not conduit.
2.
The FSAR commitments for conduit are actually in section 3.10.
3.
The FSAR damping value referenced in section 3.10 is one percent, without reference to earthquake level.
4.
The one-percent damping value contained in section 3.10 of the FSAR is out of place and was never used in design.
SQN FSAR section 3.7 (Seismic Design) lists damping values to be used for design of Category I structures, systems and components (including large and small diameter piping), but does not explicitly define damping ratios to be used for seismic support of rigid metal conduit. This resulted in the practice of treating rigid metal conduit in a fashion similar to small piping, including the use of variable damped spectra, which the FSAR allows for small piping.
The analogy to small piping was originally reflected in design criteria SQN-DC-V-13.10 as two-percent fixed damping for the safe shutdown earthquake (SSE) condition.
The design criteria were later revised to reduce the peak on the two-percent curve to correspond to the peak level for a five-percent curve. Also, the spectra used for input to the design criteria were broadened approximately 20 percent on the high-frequency side of the peak and broadened to zero hertz on the low-frequency side.
pIR SQNCEB8756 has been issued against this observation and the following corrective action is proposed. Based upon existing TVA test data, the technical bases will be developed and a summary report issued to demonstrate that the damping used for the SQN electrical conduit is conservative.
Damping has been determined from traditional, low-level excitation tests and for full-scale SSE tests for available configurations.
Furthermore, the results of tests and analyses by others will be included for comparison. Following review of the summary report, FSAR section 3.7 will be revised to add damping values for rigid metal conduit, steel, and aluminum.
Also, design criteria SQN-DC-V-13.10 will be revised as appropriate to reflect the approved damping values resulting from this work.
e ENCLOSURE 2 LIST OF NEW COMMITMENTS FOR SEQUOYAH NUCLEAR PLANT (SQN) 1.
Include the completed water hammer analysis of the containment spray system (CSS) as an appendix in the CSS calculation package by July 1, 1987.
2.
Section 6.3 of the SQN FSAR will be revised in the 1988 amendment to be consistent with the SQN SER (NUREG-0011) regarding the method of determining net positive suction head (NPSH) to reflect allowance for subcooling, 3.
During the review of unverified assumptions used in EEB calculations, EEB will revise the calculations to locate assumptions in a single dedicated section. This effort will be complete before unit 2 restart.
4.
CEB will issue a report or calculation'by July 15, 1987, to document justification.of seismic qualification for handwheel-operated globe valves of four inches and less nominal pipe size.
5.
CEB will revise SQN design criteria by December 31, 1987, to provide clarification of seismic. qualification criteria for valves and other in-line fluid system components.
_____-_ ____-_ - _ - _ -