ML20151Q338
| ML20151Q338 | |
| Person / Time | |
|---|---|
| Issue date: | 07/26/1988 |
| From: | Haass W, Potapovs U Office of Nuclear Reactor Regulation |
| To: | |
| Shared Package | |
| ML20151Q331 | List: |
| References | |
| REF-QA-99900404 NUDOCS 8808110124 | |
| Download: ML20151Q338 (17) | |
Text
.
~
ORGANIZATION: WESTINGHOUSE ELECTRIC CORPORATION PITTSBURGH, PENNSYLVANIA REPORT INS *JECTION INSPECTION N0.: 99900404/88-01 DATES:
04/18-22/88 DN-SITE HOURS-QR CORRESPONDENCE ADDRESS: Westinghouse Electric Corporation Power Systems Division ATTN: Mr. Thomas E. Campbell General Manager Post Office Box 3E5 Pittsburgh, Pennsylvania 51230 ORGANIZATIONAL CONTACT: Mr. David Alsing, Manager, Quality Assurance TELEPHONE NUMBER:
412-829-3708 NUCLEAR INDUSTRY ACTIVITY: Westinghouse provides hSSS components and services for nuclear power plants.
ASSIGNED INSPECTOR:
4 [E 7 N/I7 W. P. Haass, Special Projects Inspection Section Date (SPIS)
OTHER INSPECTOR (S):
R. L. Pettis, Jr., SPIS K. Naidu, P D APPROVED BY:
LQ M NO
- 7lle @
U. Iotapovs, Chief SPIS, Wendor Inspection Branch Date INSPECTION BASES AND SCOPE:
A.
BASES:
10 CFR Part 50 and 10 CFR Part 21.
B.
SCOPE:
Review evaluations and records reCarding several recent events at Westinghouse PWR-type nuclear power plants to determine whether their treatment was conducted and their corrective action developed and disseminated in accordance with applicable NRC regulations anJ require-ments, and review the procedures and their implementation to satisfy the requirements of 10 CFR Part 21.
PLANT SITE APPLICABILITY:
All nuclear plants with Westinghouse PWR-type NSSSs.
k h
MV 99900404
~ _ _.
s ORGANIZATION: WESTINGHOUSE ELECTRIC CORPORATION PITTSBURGH, PENNSYLVANIA REPORT INSPECTION NO.-
99900404/88-01 RESULTS:
PAGE 2 of 17 j
A.
VIOLATIONS:
~~
i
- None, j
B.
NONCONFORMANCES:
Contrary to Westinghouse procedure WRD-0PR-19.0, Revision 2, "Identifica-tion and Reporting of Substantial Safety Hazords, Significant Dericiencies, and Unreviewed Safety Questions," dated December 1, 1985 Westinghouse failed to exercise timely reporting to the NRC, until June 18, 1987, correcting a previously submitted report on July 13, 1984, relating to a potential overpressure condition in the component cooling water systems designed by Westinghouse.
C.
UNRESOLVED ITEMS:
None.
D.
STATUS OF PREVIOUS INSPECTION FINDINGS:
1.
(Closed) Violation B (84-02) - Three examples of potentially reportable issues were identified in which the evaluation records did not support the determination that the defective item was not reportable to the NRC or referable to the Safety Review Comittee (SRC). WRD-0PR-19.0, Rev. 2 contains a requirement that evalua-tions must be prepared ano retoined.
The specific records that must be included during the evaluation process are a function of the departmental procedures.
Westinghouse indicated that WRD-0PR-19.0 is the formal procedure that includes guidance for daterminir.g reportability of safety concerns in conformance with Part 21.
Repertability decisions are made by the SRC considering the specific issues that must be addre.5seo as giver in an instruction sheet to determine reportability vader 10 CFR Part 21, 50.59, and 50.55(e).
Further, there is an SRC manual that includes all the pertinent information for the SRC activities; the manual is provided only to the SRC members and conforms to the guidance given in WRD-0PR-19.0, Rev. 2.
There are no other departmental procedures addressing these activities as wa<
once contemplated by Westinghouse. Based on the review of the issues discussed in other po. ions of this report, the inspectori concluded that the procedures were properly implemented.
This matter is considered to be closed.
I
j j
ORGANIZATION: WESTINGHOUSE ELECTRIC CORPORATION PITTSBUAGH, PENNSYLVANIA REPORT INSPECTION NO.-
99900404/88-01 RESULTS:
PAGE 3 of 17 Westinghousehadnotcompleteian 2.
(Closed) UNRESOLVED ITEM evaluation of the means to accomplish on-line testing of the reactor trip interlock (P-4) and only the P-4 relay for the Safety Injection block / reset function was presently covered by a procedure.
The inability to on-line test the remaining three P-4 contacts was identified in 1985, and the objective for Potential Iteh,85-006 was expanded to address this problem.
The study was completed in 1987 with the result that only the P-4 contact which assures a trip of the main turbine when the reactor trips (to preclude sn excessive RCS cooldown event) was detennined to require testing. This infoma-tion was sent to all Westinghouse customers by letter, dated October 27, 1987, but incluied no hardware recommendations or a detailed procedure. Westinghouse considers the PI file to be closed with the affected licensees responsible for determining the need for haroware and procedures to permit on-line testability of the P-4 contact.
E.
OTHER FINDINGS AND COMMENTS:
1.
Westinghouse Policy and Procedure Regardir,g 10 CFR Part 21:
Westinghouse has issued topical report WCAP-10623, "Guidelines for Implementation of NRC Regulation 10CFR21" in July 1984, and WRD-0PR-19.0, Rev. 2, "Identification and Reporting of Substantial Safety Hazards, Significant Deficiencies, and Unreviewed Safety Questions" on December 1,1985 to specify the policy and procedures for implementation of the Part 21 requiranents. The documents adhere to the NRC requirements for reporting of defects and deficiencies in safety-related components and services including the reporting on an interim basis of such defects and deficiencies that require a long time for evaluation.
However, based on review of the several technical issues discussed in the following sections of this inspection report, it was determined that the threshold established by Westinghouse for reporting under Part 21 is significantly higher than that considered appropriate by NRC, but still within a broad interpretation of the pertinent terms in the Part 21 regulation.
In the ma,iority of instances, however, Westinghouse has initiated or performed analyses of the problem under a Potential item file, comur.icated with NRC regarding the problem by telephone and in meetings, and/or informed affected customers by lecter to assure that the generic implications were properly addressed.
In other words. Westinghouse essentially conformed to the requirements of 10 CFR Part 21 without designating the action as such.
In some of the instances reviewed, NRC deemed
08GANIZAT10N: WESTINGHOUSE ELECTRIC CORPORATION PITTSBURGH, PENNSYLVANIA REPORT INSPECTION l
NO.-
99900404/88-01 RESULTS:
PAGE 4 of 17 the informing of-other potentially affected licensees about the problem of such importance that an information notice or a bulletin was issued.
l 2.
LOCA Analysis Error at Haddam Neck: By letter dated March 25, 1988, Connecticut Yankee Atomic Power Company (CYAPC0) provided a justifica-tion for continued operation of the Haddam Neck Plant due to an apparent miscalculation by Westinghouse of the limiting single failure in the large break LOCA analysis.
CYAPC0 discovered a disparity between the limiting LPSI flow rate calculated by Westinghouse and that calculated by CYAPC0 while performing in-house analyses.
The difference in flow rate of approximately 1800 gpm less than was used in the original calculation resulted from a determinatior, that the most limiting flow condition involves the failure of a LPSI valve which would result in flow through two penetrations rather the four provided.
The original analysis was based on the failure of an emergency diesel generator that deenergi:es a train of safety injection.
The net effect of this error identification was a reduction in the allowable peak linear heat generation rate from 14.3 to 13.3 kw/f t to assure that the peak clad temperature would not exceed 2300 degrees F.
Westinghouse opened a Potential Item file to verify the CYAPC0 findings, to conduct a rev4w of all other Westinghouse PWRs (even though Haddam Neck is unique in tH " regard) to assure that this problem does not affect other planu (especially those with low flow resistance where the loss of an injoction path causes a high restriction on flow), and to identify the cause of the original erroneous conclusion.
At the time of the inspection, no schedule hao as yet been established for completion of the PI file. No Part 21 report was issued.
l A search of the files by Westinghouse could not locate a copy of the l
original calculation performed in 1972.
However, the inspector l
reviewed another calculation produced in the same time frame and determined that appropriate verification centrols were applied in aCCordance With 10 CFR Appendix B.
The calculations for the current analysis were also reviewed; the analysis utilized the PEGISYS code and verification was performed.
The analysis confirmed that the CYAPC0 results were reasonable and conservative.
3.
Thimble Tube Wear:
Thimble tube wear was first reported by a domestic plant (Salem 1) in 1981 and by a French nuclear plant j
(Palual) in 1985. Several donestic nuclear plants have since reported excessive wear in the retractable thimble tubes that house
ORGANIZATION: WESTINGHOUSE ELECTRIC CORPORATION PITTSBURGH, PENNSYLVANIA REPORT INSPECTION NO.-
99900404/88-01 RESULTS:
PAGE 5 of 17 the moveable in-core detectors utilized in Westinghouse PW E. The wear occurs in the fom of denting and thinning of the walls of thimble tubes and through-wall cracking of the central guide tube in the fuel assembly, at the elevation of the lower core support structure primarily due to flow-induced vibration.
The problem seems to manifest itself primarily on 14 foot cores, but has also been seen on 8 and 12 foot cores. The safety significance of most importance involves the loss of RCS pressure boundary integrity in the event of a thimble tube rupture.
Westinghouse does not consider thimble tube wear and possible rupture to be a safety issue because on analysis of a single tube failure (leakage area is 0.00024 feet) at a RCS pressure of 2250 psia produces a leakage rate that can be accommodated by a single charging pump.
In fact, the leakage from three tube failures of the size can be maae up by a single charging pump.
(Note that Whnical Specifica-tion requirements demand shutdown due to the lc uage from one tube rupture.) Further, all domestic plants have manually operated valving except for Callaway and Millstone 3 (the system for San Onofre 1 was not provided by Westinghouse) and can therefore isolate leakers as they occur. Westinghouse has not established a PI file for this issue nor has it issued a Part 21 report.
NRC issuad Information Notice 87-44 in September 1987 and EPRI issued Plant Infomation Notice 87-01 on this subject.
A generic information letter was issued by Westinghouse to all its customers in November 1986 to indicate the need for awareness of this potential problem since thimble tube wear was observed on some 12-foot cores (Salem and Ringhals) and tube failure could result in a small break LOCA.
The most rapid thimble tube wear is occurring on plants with 14-foot cores. The South Texas unit, currently undergoing startup testing, has been eddy current tested during an interim shutdown and found to have 19 thimble tubes with wall thinning ranging from 12 to 58.7 percent.
The Tahange unit had tube wear-through in three months.
Doel four had two tubes wear through af ter 10 months of operation, 32 tubes exhibit wall thinning of greater than 88 percent, and an additional 47 tubes exhibit wall thinning of 35 to 88 percent. Of the 12-foot cores, Salem appears to be unique showing excessive tube wear within three refueling cycles. Approximately 20 nuclear units of this size are known to have detected wear to some degree but are j
expected to be operable without excessive wear fc" 10 to 15 years.
Westinghouse recommended corrective action to date involves the installation of flow limiting devices, which have not been very successful, and relocating the position of the tubes to absorb the w
ORGANIZ'ATION: WESTINGHOUSE ELECTRIC CORPORATION PITTSBURGH, PENNSYLVANIA REPORT INSPECTION NO.-
99900404/88-01 RESULTS:
PAGE 6 of 17
~
(
vibrational wear at a different point on the tube. Attempts to resolve this issue have not been successful to date and have taken l
considerable time.
l 4.
Failure of W-2 Type Cell Switches:
On May 15, 1987, a problem was i
ioentified with a W-2 cell switch at the Indian Point Unit 3 nuclear f
plant. The W-2 switch is located in a 480 volt circuit breaker (CB) l cell and provides an input to the appropriate logic as to whether I
the CB is racked in or out.
The W-2 switch used in a 480 volt CB cell is a modified version of the W-2 control switch. At Indian Point 3, the W-2 switch provided l
input to the Emergency Diesel Generator (EDG) logic system to permit j
the operation of the EDG output CB depending on the position of the CB.
Improper functioning of the W-2 switch will provide an erroneoes input to the EDG logic system which in turn will prevent the EDG l
output CB from closing.
1 Westinghouse opened a Potential Item File (PI-87-038), evaluated the failure mode, and notified the NRC on October 16, 1987, pursuant to the requirements of 10 CFR Part 21.
When the CB is racked in, the CB presses a lever on the W-2 switch thereby providing an appropriate indication to the logic. When the CB is racked out, the lever should return to its nonnal position thereby providing that indication to the logic.
It was concluded from analysis and evaluation of the failed switch that age-related deformation of the spring retainer in the spring-return mechanism was the root cause of the erroneous l
indication. As a result, NRC issued Information Notice 87-61 to inform licensees of the potential W-2 switch problem, and to convey the Westinghouse recomendation that visual verification of the operating position of the lever should be conducted to assure opera-bility of the W-2 switch.
On March 23, 1988, Westinghouse notifiec' the NRC in a supplement to the original Part 21 report that visual inspection of the operating lever position was not a reliable indica-l tion of the W-2 switch operability.
Westinghouse now recommends I
that switch operability be detdrmined by an electrical continuity l
check using a spare set of contacts whenever the CB is removed from l
its "connected" position.
NRC issued Supplement 1 to the information l
notice on May 31, 1988 conveying this new recommendation.
5.
Potential Overpressurization of the CCWS: On June 18, 1987 Westing-bouse notified'the NRC, pursuant to the provisions of 10 CFR 21, of j
a reportable matter associated with prev 1ously recommended design modifications to Westinghouse supplied Component Cooling Water Systems (CCWS) for the Point Beach 1 & 2 nuclear plant.
The notification j
0.RGANIZATION: WESTINGHOUSE ELECTRIC CORPORATION PITTSBURGH, PENNSYLVANIA REPCRT INSPECTION NO.-
99900404/88-01 RESULTS:
PAGE 7 of 17 was also sent to other affected plants including Turkey Point 3 & 4 H. B. Robinson, Ginna, and Kewaunee, as well as other potentially affected plants for information only. The matter related to an earlier Part 21 report dateo July 13, 1984 in which Westinghouse recommended certain plant modifications be implemented to resolve a potential overpressure condition in the CCWS which could result from closure of the surge tank vent valve on a high radiation signal from the radiation detectors located within the CCWS.
Closure of the vent valve could result in an increase in pressure (above the normal atmospheric pressure) in the surge tank due to a system inleakage or an increase in system heat load. The surge tank pressure could then increase to the set pressure of the relief valve which, in combination with the developed head of the CCW pumps, would create an over-pressurization condition of up to 170 percent of design pressure downstream of the CCW pumps.
To alleviate this potential problem.
Westinghouse recommended:
(1) disabling the circuitry which closes the CCWS surge tank vent valve and converting the valve to a locked open local valve, and (2) removing the surge tank relief valve or its internals. Westinghouse stated that these measures effectively achieve a low pressure drop overflow path from the surge tank which would ensure that the maximum pressure in the CCWS would not exceed 110 percent of the design pressure in the event of operation with a water solid surge tank coincident with the maximum anticipated inieokage through a ruptured tube in one of the system heat exchangers.
Typically, the limiting condition is the rupture of a tube in the reactor coolant pump thermal barrier heat exchanger resulting in system inleakage equivalent to 260 gpm under cold conditions.
However, additional review by Westinghouse, prompted by a response to the original recommendations from Point Beach, indicated that implementation of these recommendations would make the CCWS an open system outside of containment. This would be a violation of contain-ment isolation requirements for the plants listed as described in their FSARs, since their dcsign basis may depend on the CCWS outside of contoinment being a closed system and serving as an extension of the containment boundary.
In February 1985, a Westinghouse internal letter identified the need to review reissuing the 1984 Part 21 letters to the Point Beach, Ginna, and Kewaunee licensees to reassess the validity of the original recommendations.
During the inspection, Westinghouse could only locate two reissued Part 21 letters, one to Point Beach, dated April 24, 1985, and the other to Kewaunee, dated August 14, 1985,
ORGANIZ'ATION:
WESTINGHOUSE ELECTRIC CORPORATION PITTSBURGH, PENHSYLVAhlA i
REPORT INSPECTION NO.-
99900404/88-01 RESULTS:
PAGE 8 of 17
~
identifying the fact that the earlier recomendations, if acted upon, may result in the violation of plant containment isolation commitments.
Based on the original recomendations, Kewaunee had removed the relief valve internals in July 1984, which mode their CCWS an open-system as described above ind had a potential for violdting their containment isolation commitments. The revised Westinghouse recomendations offered two options for consideration.
The first was to reinstall the relief valve internals and justify by analysis that the CCWS components and piping can accomodate the higher pressures previously mentioned.
The second option was to modify the FSAR to incorporate the system design changes.
In a 10 CFR 50.73 report, dated July 20, 1987, Kewaunee stated to the NRC that evaluation of the June 18, 1987 Part 21 report was begun immediately with the assistance of both Westinghouse and Fluor Engineers, Inc. (the original architect-engineer) with the result that the Part 21 notification did not apply to Kewaunee and therefore no corrective action was required.
The basis for this conclusion was that the modified CCWS configuration continues to meet the closed system requirements inside containment thereby providing one of the two required barriers. This was verified by comparing the CCWS design to the nine criterie for a closed system inside containment as outlined in Section 3.5 of ANSI /ANS 56.2-1984.
Information was unavailable as to what actions, if any, Kewaunee initiated earlier as a result of Westinghouse's August 14, 1985 letter.
Westinghouse determined that this problem potentially existed for those vintage plants designed prior to the issuance of General Design Criteria 56 and 57, since plants designed after that time would not have taken credit for a closed system outside containment as one of the redundant isolation features.
The inspectors reviewed Potantial Item File 84-269 which was opened by Westinghouse to evoluate this matter.
The inspectors concluded that the occumentation reasonably indicated that Westinghouse possessed sufficient information in 1985 to issue notification to their customers and the NRC that the original recomendations were erroneous. The file included a memorandum dated February 1986 in which the need to issue a revised Part 21 report to the NRC was identiti. d. and considered the possibility of issuing a Westinghouse Technical Bulletin. On June 16, 1987, Westinghouse convened their internal Safety Review Committee to address the issue of reportability.
It was concluded that this issue should have been reported to the NRC and Westinghouse customers in 1985 so that appropriate corrective I
- l ORGANIZATION: WESTINGHOUSE ELECTRIC CORPORATION PITTSBURGH, PENNSYLVANIA REPORT INSPECTION NO.-
99900404/88-01 RESULTS:
PAGE 9 of 17 action could have been taken. Westinghouse notified NRC on-June 18, 1987, more than two years subsequent to possession of sufficient infomation regarding the erroneous nature of the recommendations previously disseminated.
This is the basis of the nonconformance described in an earlier part of this inspection report.
6.
Minimum 3H of Containment Spray Fluid:
On September 16, 1987, Georgia Power Company, licensee for the Vogtle nuclear plant, reported that a reanalysis determined that, in the event of a postulated large break LOCA, the equalizing sump solution for the Containment Spray System (CSS) may not achieve the stated design basis minimum pH of 8.5 to assure retention of iodine in solution.
Using the most conservative assumptions, the minimum pH was calculated to be 8.1.
The CSS is provided to reduce containment pressure in the event of a LOCA; it alsc removes iodine from the containment atmosphere and adjusts the rump solution pH within prescribed limits. Adjustment of the sump solution pH between 8.5 - 10.5 assures retention of iodine in solution and limits chloride induced stress corrosion cracking of stainless steel.
The environmental qualification of equipment installed in the areas sprayed may be adversely affected if the pH exceeds 10.5.
The CSS pumps take suction from the Refueling Water Storage Tank and mix it with sodium hydroxide in the Spray Additive Tank (SAT) for pH control during the injection mode; pump suction switches over to the containment sump for the recircula-tion mode.
The matter of obtaining the proper decontamination factor (DF) for iodine retention in the sump was first identified at the South Texas Project (51P) nuclear plant on February 9, 1984 where the DF was detercined to be nonconservative. Westinghouse opened Pottntial item File 84-245 to acoress the issue and determine the proper corrective action. The matter was reported to the NRC by the licensee as a 50.55(e) item. The corrective action detemined for STP to be in compliance with FSAR commitments was to eliminate the spray additive during the recirculation mode. This resulted in a reduced equilibrium sump pH of 7.5 and an acceptable revised offsite iodine dose. The pH value of 7.5 in the sump exceeds the minimum value of 7.0 specified in Branch Technical Position MTEB 6-1 necessary for the protection of r.ainless steel equipment.
STP nctified NRC Region IV in a letter dated September 8, 1986 of this resolution to the 50.55(e) item.
The Westinghouse analysis also determined that the Farley and Vogtle nuclear plants were possibly adversely affected by this concern.
]
ORGANIZATION: WESTINGHOUSE ELECTRIC CORPORATION PITTSBURGH, PENNSYLVANIA REPORT INSPECTION NO.-
99900404/88-01 RESULTS:
PAGE 10 of 17 On September 15, 1987 Westinghouse informeo the Farley facility by letter thot in the event of 4 postulated large break LOCA, the CSS may not achieve the minimum equilibriurr sump sclution pH of 8.5 which is the stated design and licensing basis.
Therefore, an unreviewed safety question as defined in 10 CFR Part 50.59 existed.
As a result, Forley provided additional information regarding their current operating practices. Westinghouse reviewed this information and provided recommendations to revise their operating procedures to assure a sump pH consistent with licensing requireathts.
For the Vogtle problem, Westinghouse obtained additional information from Vogtle regarding the specific operating procedures and recalcu-lated the equilibrium sump solution pH to be 8.58.
The Vogtle operating procedures were revised to ensure that the SAT is isolated before switching to the recirculation mode from the injection mode.
On September 19, 1987, Vogtle informed the NRC that they ware notified by Westinghouse that incorrect assumptions about the plant were used in the original calculations and that revised calculations indicate a pH value cunsistent with the plant Technical Specifications. The Potential Item file has been closed.
It appears that the resolution of this matter could have been accomplished more expeditiously.
7.
Nonvital Power Supply Used in Valve Interlock Logic: On December 2, i
1987, Carolina Power and Light Company (CPL) notified the NRC that its Robinson facility had a potential single failure criterion problem.
The problem was identified as > result of a Westinghouse letter to CPL dated November 3, 1987 and involves the possibility of two redundant valves failing to open during the shifting of the safety 1
injection system from the injection mode to the recirculation mode within three minutes following initiation of a LOCA. One of the valves is required to open in order to successfully switch from the
)
injection mode to the recirculation mode. An interlock is provided to protect the lower design pressure residual heat removal (RHR) piping from overpressurization when connected to the reactor coolant system.
In addition, the interlock prevents backflow to the refueling 1
water storage tank.
The problem arises due to the fact that the interlock is powered from c non-vital ir.strument power supply (NIPS) which is not energized during an accident if offsite power is also lost. A single failure of the NIPS could disable the interlock circuitry and consequently would prevent both valves from operating.
1 The NRC was notified of a similar problem by Florida Power and Light Company at their Turkey Point nuclear plant by a letter dated July 23, 1984.
Licensee Event Report, dated July 24, 1984, was enclosed.
An
ORGANIZ'ATION: WESTINGH0USE ELECTRIC CORPORATION PITTSBURGH, PENNSYLVANIA REPORT INSPECTION NO.-
99900404/88-01 HESULTS:
pAGE 11 of 17
~
internal Westinghouse memorandum, dated July 17, 1984, indicated that the licensee requested Westinghouse to evaluate this problem on July 13, 1984. Westinghouse opened Potential Item File 84-271 to adoress thfs matter. The evaluation included the review of generic implications for other Westinghouse nuclear plants. However, it was determined that the design of power distribution systems for nuclear plants is not within the scope of the NSSS design ceveloped by Westinghouse; this is an architect-engineer responsibility.
Westinghouse did attempt to obtain additional information on this matter but was unsuccessful.
In November 1987, Westinghouse issued d Technical Bulletin to its customers informing them of this potential problem. As a result of this notification, CPL evaluated the design for Robinson, identified a pr,tential problem, and notified the NRC. Again, it appears that the Westinghouse resolution to the problem could have been accomplished in a more expeditious manner.
8.
Fuel Rack Hold Down Bolts: On November 27, 1987, Houston Lighting and Power Company (HL&P) notified the NRC pursuant to 10 CFR Part 21 that the spent fuel rack hold down bolts at their South Texas nuclear plant were found significantly undertensicned.
The bolts that anchor the fuel racks to the embedments require a pretension force of 33,000 pounds which is equivalent to a torque of 1100 ft-lb as stipulated in Westinghouse Design Calculation WNEP-7816, dated August 1978. The dasign and manufacture of the spent fuel racks and their mounting requirements are the responsibility of Westinghouse Nuclear Components Division (NCD) located in Pensacola, Florida.
During field erection ano installation of the fuel racks, the embedment stud / nut torque value required by Westinghouse procedure 2463 A82(1978) could not be attained.
It was verified by the NRC inspector on site that o torque of 25 ft-lb was used instead of the specified 1100 ft-lb. A revised design calculation was performed by Westinghouse using the 25 ft-lb torque value on the stud / nut. This cair.ulation demonstrated acceptable loadings with a safety factor of 1.58.
The Westinghouse NCD engineers' position with respect to bolt preload was that the governing loading condition for the support pads / bolts is horizontal seismic, which causes a berding moment and shear force on the base of the rack and causes the support pads / bolts to resist shear as well as tension on one side and compression on the other side. The previous analysis was based on use of the bolt preloads on the support pads so that the shear force at the rack base could be resisted by all four supports. The revised analysis (TR Book 205, pp 140-146) assumed preload resulting from 25 f t-lb i
ORGANIZATION: WESTINGHOUSE ELECTRIC CORPORATION PITTSBURGH, PENNSYLVANIA REPORT INSPECTION M0.-
99900404/88-01 RESULTS:
PAGE 12 of
.7 bolt torque and resisting the base moment with two bolts in tension and two pads in compression with all the shear force applied to the two compression pads.
The installation procedure called for use af the "Turn of the Nut" method and a bolt / nut torque of 1100 ft-lb.
This method would achieve a bolt preload of 33,000 pounds / bolt. However, the revised ar.alysis did not take credit for the bolt preload since it was not needed at all.
Corrective actions taken by HL&P included a review of all other Westinghouse field erected equipment installation procedures including fuel handling machine, refueling machine, fuel transfer system, and the reactor in' m als.
This review determined that the torque valum for bolted je requiring preload have. been specified in the installation r.ucedure as required by design calculations.
- Also, the "Turn of the Nut" method was not used in the procedure.
HL&P Engineering reviewed the Westinghouse reanalysis calculations and has verified that an adequate margin of safety exists. This analysis is not applicable to the racks in the spent fuel storage pool since they are designed to be 14-inch center-to-center free standing racks for which no anchor bolting is required. As a result, Westinghouse WCD issued Field Deviation Report THXM-11058, dated November 30, 1987, which accepted the revised calculation for the installed condition. Westinghouse stated tiiat to their knowledge no other plant utilizes these in-containment bolted racks; other plants utilize the free standing type acks whose design is based on a non-linear structural analysis.
9.
RHRS Pump Miniflow Requirements: Two design concerns have been identified on RHRS pumps regarding potential damage and failure due to considerable testing and operation at low flow conditions. One concern involves the possible deadheading of a pump in arrangements in which two or more pumps are,in parallel and nne pump is stronger than the other.
The other concern involves the provision of insufficient flow capacity in bypass lines used during pump testing and while awaiting RCS pressure degradation, insufficient miniflow can result in pump overheating, excessive impeller / shaft vibration, excessive forces on the impeller, and cavitation leading to foilure.
It is not a problem recognized Dy the pump manufacturers since it is probably uniqi.e to nuclear power plants due to the considerable testing and operatiun at low flow conditions.
ORGANIZATION: WESTINGHOUSE ELECTRIC CORPORATION PITTSBURGH, PENNSYLVANIA REPORT INSPECTION N0.: 99900404/88-01 RESULTS:
PAGE 13 of 17 This problem was initially identified at Turkey Point in early 1987.
Westinghouse openea Potential Item File 87-020 to study the issue and determine what corrective action would be necessary. Westinghouse issued a letter to its affected customers (13 Units in 9 plants), dated October 26, 1987, informing them of the problem and recomending that each utility review its safety injection miniflow configurations and pump manufacturer's miniflow requirements to determine applica-bility of the concerns.
It further indicated that thc. concerns if applicable would require prompt operator action in certain accident situations to preclude pump damage and thereby assure availability for accident mitigation. An& lysis indicates that pumps can operate no longer than 10.4 minutes in a deadheaded condition without overheating at which point administrative procedures could be relied upon to require pump shutdown.
A followup generic letter was issued by Westinghouse that included further possible actions by utilities. It recomended that all ECCS pemps be tested to determine whether deadheading is occurring.
It suggested a review of performance history to identify whether the pumps have been operated using the minifiow lines alone, and to verify pump performance to determine if degradation has occurred.
It recorrnended that the Emergency Operating Procedures be modified to incorporate the administrative controls and that the ability for proper conduct of operator action be verified.
Further, PRA analyses should be performed to determine frequencies of initiating events j
for actuation of S1 and the increase of core damage frequency if a LPSI pump faileo due to deadheading.
It was indicated that the ideal solution is to provide individual minimum recirculation lines of proper capacity for each affected pump.
Westinghouse did not issue a Part 21 report citing a lack of safety significance since a very minimum flow will provide some protection against damage, insufficient piping system information was available to make clear d; terminations, and no incident of pump damage due to deacheading or low flow could be confinned. Westinghouse did consider the problem to be reportable by licensees under 10 CFR 50.59 as an unreviewed safety concern since the FSAR considered loss of a pump on system startup and not loss of a pump due to deadheading or insufficent miniflow capacity.
Westinghouse has also held telephonic oiscussions cn these concerns with cognizant NRC personnel. On this basis, Westinghouse closed the PI file on December 14, 1987.
NRC issued Dulletin No. 88-04 "Potential Safety-Related Pump Loss,"
dated May 5,1988, notifying all licensees of this problem and requesting an evaluation of applicability to their facility and the corrective action planned to resolve the problem.
ORGANIZATION:
WESTINGHOUSE Ei.ECTRIC CORPORATION PITTSBURGH, PENNSYLVANIA REPORT INSPECTION NO.-
99900404/88-01 RESULTS:
pAGE 14 of 17 Westinghouse has decided to open another PI file shortly to address as a separate issue those plants that have individual miniflow lines without check valves to preclude pump-to-pump interaction. This group consists of 13 Units in 7 plants.
10.
Steam Generator Tube Rupture:
On July 15, 1987, a steam generator tube rupture event occurred at North Anna Unit I while at 100 percent power. The leakage location was found to be at the top tube support plate on the cold leg side of the tube.
The opening was circumferen-tial and extended 360 degrees around the tube. The cause of the tube rupture was determined to be high cycle fatigue due to a combination of loads consisting of mean stress produced by denting of the tube at the top tube support plate and the alternating stress due to out-of-plane deflection of the tube above the top tube support caused by flow induced vibration.
Denting of the tubes resulted from the presence of corrosion products in the tube-to-support plate crevices in carbon steel support plates, and flow induced vibration resulted from the lack of an antivibration bar (AVB) support.
Westinghouse opened PI File 87-033 on September 17, 1987 to address the generic implications and corrective action for this problem.
On September 22, 1987 Westinghouse issued letters to its utility customers known to be affected (Type I) to advise them on the potential for a tube rupture event, to recommend evaluation of primary to secondary leak rate determination measures to assure rapid detection of a tube rupture event, and to offer Westinghouse assistance in resolving this problem.
At t e same time, Westinghouse also issued d sir.ilar letter to its other customers beliesed to be unaffected (Type II), as a precautionary measure, to reevaluate their situation to assure that no tube denting exists, and to offer Westinghouse assistance as needed.
The Westinghouse analysis indicated that generally a steam generator conta %s zero, one, or two tubes whose bend is located below the effective support provida.d by the AVBs, and with higher than average secondary flow past the tube due to the restriction to flow caused by nearby AVBs. This arrangement results in flow induced vibration which, combined with increased stress caused by the denting of the tube at the carbon steel tube support plate due to corrosion products, can lead to tube rupture Generally, i nuclear unit will contain six s'asceptible tubes.
The analysis also considered the potential for a combined tube rupture and a steamline break which was found to be extremely small.
~
1 ORGANIZATION: h2STINGH0VSE ELECTRIC CORPORATION PITTSBUkGH, PENNSYLVANIA REPORT INSPECTION H0.-
99900404/88-01 RESULTS:
PAGE 15 of 17,
~'
Based on the above, Westinghouse concluded that fatigue cracking and subsequent rupture of a steam generator tube does not represent a substantial safety hanrd reportable under 10 CFR Part 21.
Rapia crack propagation is not expected and leakage, should it occur, is expected to result in offsite dose rates well within the limits of 10 CFR Part 100.
Further, multiple tube ruptures is considered a low probability event sad a single tube rupture is a design basis event.
The PI file was closed.
On February 5, 1988, NRC issued Bulletin No. 88-02, "Rapidly Propagating Fatigue Cracks in Steam Geanerator Tubes," to all holders of operating licenses and construction permits for Westinghouse and Combustion Engineering designed nuclear power plants to implement actions to minimize the potential for a steam generator tube rupture event caused by a rapidly propagating fatigue crack.
11.
Inaccurate Power Range NI for Lor: Leakage Cores:
On February 6, 1988, operators at M.llstone Unit 3 experienEd difficulties in aojustment of the power range (PR) and intermediate rarge (IR) nuclear instrumentation (NI) during startup after the first refueling outage.
Specifically, the gain circuit in the PR NI did nct permit adjustment of the PR instruments to match the actual thermal power with a reactor core designed for low neutron lee'< age as at Millstone Unit 3.
prs are calibrated to indicete the calculated therm:1 power of the reactor.
Improper calibration cay result in reactor trips.
The transition to low leakage cores requires hardware modifications to the gain circuits in f.he prs to enhance the sensitivity of the PR to detect a reduced level of neutror *ctivity at the detector location.
Low leakage cores are used to decrease the rate of neutron embrittle-ment of the reactor pressure vessel and increase fuel economy.
One of the effects of a low leakage core is reduced neutron leakage along the entire core periphery.
However, the reduction is not uniform and affects the prs more than the irs due to their physical location.
Other factors which affect tre IR and PR NI are the age cf the instruments and their sensitivity to respand to the dec Tase in neutron leakage. Westinghouse also supplies low leakage reload cores to other than Westinghouse-designed NSSSs.
The Westinghouse Nuclear Fuels Division (NFD) is the (sgnizant division responsible for problems related to nuclear fLels. NFD was aware of similar prob M s at Zion ed DC Cook prict to Millstone 3.
In response to the preblem identified at Millstone 3. Westinghouse issueo a field changa nctice (FCN) for the installation of resistors and potentiometers in the PR.NI gain circuits W increase the adjust-
=. -
ORGANIZATION: WESTINGHOUSE ELECTRIC CORPORATION PITISBURGH, PENNSYLVAhlA REPORT INSPECTION NO.-
99900404/88-01 RESULTS:
PAGE 16 of 17 ment band.
This FCN was similar to the one used at DC Cook-in 1983 when the same problem was encountered with a Westinghouse low leakage reload core.
In a letter dated September 16, 1983 to the DC Cook licensee, Westinghouse provided specific guidance for the adjustment of IR and PR NI because of the transition to a low leakage core.
The letter stated that the guidance arose from Westinghouse's assessment of an LER issued by Zion and a subsequent INP0 Significant Event Report (72-82) concerning an inadvertant reactor trip caused by a mismatch in the IR and PR NI precipitated in part by a low leakage core.
The Westinghouse NFD representatives stated that they were aware of the difficulties encountered at Millstone 3 in January 1988 during the cycle 2 reload of a low leakage core.
Corrective action imple-mented by NFD was the development of Procedure 5.22, "Nuclear Instrumentation System Pre-Alignment Transmittal," dated February i
1988. This procedure alerts the NFD desicaers to the importance of pre-aligning the IR and PR NI prior to the initial criticality of a reload core. The procedure requires the NFD designer to transmit l
information necessary to the customer plant operating personnel in a i
timely manner so that the expected impact of the core design changes on detector output currents can be evaluated.
This information would permit re-alignment of the IR and PR NI prior to the initial l
criticality of the reload core. As of March 1983, Westinghouse has l
sent letters to their customers providing guidance on the alignment of the IR and PR NI prior to reload startup and to evaluate the need l
for potential hardware modifications to the PR N! due to the minimum I
sensitivity of the PR detector electronics. Westinghouse proposes to issue a technical bulletin on this subject in the near future. No Potential item File was opened on this issue.
E.
PERSONS CONTACTED:
Westinghouse NTSD:
i it David N. Alsing, Quality Engineering Menager
- + Robert A. Wiesemann, Manager, Regulatory and Legislative Affairs
- + Hedy Abromovitz, Manager, Product Assurance l
4 Walter D. Tauche, Manager, Safeguards Analysis i
- + Brian A. McIntyre, Manager, Product Licensing
- + William J. Johnson, Manager, Nuclear Safety i+ Raymond M. Tajc, Senior Engineer, Product Assurance I
t+ Michael H. Shannon, Manager, Plant & Systems Evaluation Licensing j
f+ Carl W. Hirst, Manager, RCS Components Licensing
OBGANIZATION: WESTINGHOUSE ELECTRIC CORPORATION PITTSBURGH, PENNSYLVANIA l
REPORT INSPECTION NO.-
99900404/88-01 RESULTS:
PAGE 17 of 17
- + Samuel L. Cunningham, Principal Engineer, Quality Assurance --
- + Ernest K. Figenbaum, Principal Engineer, Nuclear Safety Coordinator
- Peter J. Morris, Managsc,1 & C System Licensing
+ Jack Hammond, Product Assurance and Support Operations
+ John D. McAdoo, Assistant Manager, Nuclear Safety
+ Beth Hall, Engineer, Licensing
+ Ronald P. DiPiazza, Manager, Operating Plant Support John S. Galembush Dan Merkovsky David E. Boyle Ed C. Arnold Paul Beczak Kathy J. King Bob Sterdes Regina K. Stirzel J. L. Gallagher, General Manager, Nuclear Technology Division Gary W. Whiteman Tom Pitterle Dick B. Miller Nuclear Regulatory Commission-
- + Walter P. Haass, hRR/VIB
+ Uldis Potapovs. Section Chief, NRR/VIB
+ Robert L. Pettis, Jr, NRR/VIB
- + Kamal R. Naidu, NRR/VIB
- - Attended Entrance Meeting
+ - Attended Exit Meeting