ML20245H723
| ML20245H723 | |
| Person / Time | |
|---|---|
| Site: | Limerick  |
| Issue date: | 06/21/1989 |
| From: | Hunger G PECO ENERGY CO., (FORMERLY PHILADELPHIA ELECTRIC |
| To: | NRC OFFICE OF INFORMATION RESOURCES MANAGEMENT (IRM) |
| References | |
| NUDOCS 8906300065 | |
| Download: ML20245H723 (16) | |
Text
,
- 9s 3 'i.
- ..g,..[ -
'1 r.
. PHILADELPHIA A ELECTRIC COM PANY-Q
' NUCLEAR GROUP HEADQUARTERS
' 955-65 CHESTERBROOK BLVD, WAYNE, PA 19087 5691 (215) 640-6000 June 21, 1989 U.S.' Nuclear Regulatory Commission Attn:
Document Control' Desk Washington,-DC. 120555 Docket No. 50-353 Construction Permit No.'CPPR-107 t
Subject:
Limerick Generating Ft3 tion, Unit 2 Independent Construction Assessment Inspection Report 50-353/8.'-200
Reference:
Letter'from S. A. Varga (NRC) to C.-A. Hunger, Jr. (PECo) dated May 17, 1989 Gentlemen:
Your letter of May 17, 1989, forwarded the results of the Nuclear Regulatory Commission (NRC) inspection of corrective actions taken by' Philadelphia Electric Company (PECo) in response to' deficiencies identified during conduct of the Independent Construction Assessment being performed on Limerick Generating j
Station, Unit 2 (LGS-2).
o
. Attachment 1 to this letter provides the PECo.
J responses to items identified in the inspection report as requiring additional information for NRC resolution.
Additionally, based on re-examination of the information and root causes provided in our responses to Observation Reports and NRC Inspecti'on. Report 50-353/88-202 concerns, we are providing 1
revised or supplemental information to clarify our responses to the issues identified in Section 4 of this inspection report (Attachment 2).
Due to delays encountered in preparing this response, it is being submitted two days later than the requested submission date.
We apologize for any inconvenience this delay may have caused.
i ggO6300gcK 05000353 65 890621 g
AD G
p m
PNU e
If you have'any' questions or require any additional information, please let me know.
1 Sincerely,
. 0<
if.
.j G.
A.
Hunger, Jr.
Director, Licensing Section Nuclear Support Division Attachments i
Copy to:
W.
T.
Russell, Region I Administrator, USNRC T.
J.
Kenny, LGS Senior Resident Inspector, LGS l
1 I
l l
i l
1 i
t
.. o s
~
4
~'
1.: NRC'.OPENLITEM' 'Section-4.4.2 --Qualification of Motor-t
/ Operated. Valve InternalsL T
Concern
- The-NRC' team identified' concerns 1regarding the'use of Belden
'l
'J type AWM crosslink wire in valve actuator HV-52-2F001C.
Specifically, the motor leads ~ appeared to be. undersized for-the' application, and; evidence of their environmental qualification ~was not-provided by the licensee.
Immed'ately following the inspection, the licensee-provided; i
information that the. wire's continuous' current. rating (18 amperes) was greater than.the motor's locked-rotor rating (15.8/ amperes).. However, the wire rating was' based onlan-ambient temperature ofl30-degrees Celsius, and the wire'c rating at a more realistic ambient temperature of 40 degrees Celsius would only-be 16 amperes.
To resolve this; concern fully,'the'NRC team requires additional informationLregarding~
the'Belden wire's fire retardency, its short-term current rating-, and.the relationship.of the wire's' ratings to the design requirements of the motor-operated valve.
Response
Valve HV52-2F001C is the core spray pump suction prima'ry containment isolation ~ valve.
The motor. operator on this valve is, 1 HP with a' full load current draw of 2.6 amps and locked rotor current of 15.8 amps.
The Belden J.18 AWG are made of flame retardant cross-link polyethylene which have undergone horizontal-flame testing.
This wire is rated at'18 amps'(continuous - in free' air, approximately 90-degrees C
. based on 30 degrees C ambient temperature).
L.
The pigtails are rated for 125-degrees C.
The ambient-temperature in the core spray isolation valve compartment during normal operation.is 65 - 115 degrees F with a' maximum J
temperature of 120 degrees F.
The actual temperature reading L,
at the core spray pump room-is approximately.80 degrees'F u
(average value).
Therefore, the wire has adequate ampacity considering a maximum ambient temperature of 120 degrees F and'a locked rotor current of 15.8 amps.
]
The ampacity is determined by the insulation rating which is l
125 degrees C.
Therefore #18 AWG Belden wire is adequate.
Based on the motor load current requirements and cable ampacity for #18 AWG used for motor leads, we conclude that vendor supplied power leads are sized adequately to carry the motor current in all operating modes of the motor.
l l
Q lc)f:ls W '.'t s
LAttachmnntilh C
u
- l2. :NRC OPEN' ITEM'- Section.4.6 - ICA Trend Analysis
' Concern'
identified trend involving alteration of electrical equipment 1 after initial installation land quality control inspection.
The)Bechtel; investigation-in. response to'the'CORLfound, however, that. the ; equipment problems'were generally due :to -
~
H Lincomplete performance of the' initial installation quality verifications.
The NRC team reviewedithe CORs'related'to D'
COR-56 and'the Bechtel analysistof:the perceived trend.1
' Based upon the review,-the team concurred:with the Bechtel
- response that the' discrepancies'were generally not' caused by-n
' post-inspection work: activities..
Based upon a review of concerns identified by.both SWECLand-the'NRC, the. team. detected.a. trend associated with improperly performed. quality control inspections (89-200-01).
Although the-individual deficiencies would not have adversely; affected the' safe / operation of the-plant, _and althoughLthe
. corrective. actions-implemented by the licensee have: generally:
resolved the concerns, these issues are still indicative that the installation. quality' verification process failedito ensure. fully.that,Lin these instances,-equipment was installediin accordance with design. requirements.,BasedLupon-the scope of theJSWEC and.NRC. inspections, the team' concluded:
'that;these concerns did not constitute a programmatic-A breakdown since.the. majority of. equipment was proper 1'y installed.
Response
The' Limerick Quality' Program is totally integrated into every aspect of the project..The Quality Program begins with the
~
design.and continues through the fabrication and installation
-process.
The Quality Program is-an important,part of the
'Preoperational Testing Phase.
The primary design responsibility for the Limerick Project rests with General Electric (GE) for the Nuclear Steam Supply
. System and Bechtel for the remainder of the plant.
Both companies'have established a design process that ensures the l
Edesign concepts are properly translated into specifications, l
drawings and installation instructions.
General Electric and d
Bechtel have established procedures that have been used for I
the design process of the Limerick project.
The procedures require control of all aspects of the design.
-l J..
r '
a
-One.of the most.important elements of the design process is the series of design reviews-that have been established.
The reviews are performed by individuals independent of the initial design to ensure that the output documents properly reflect the' design.
The design output documents are used during the construction phase of the project by Bechtel.
Bechtel is also the constructor for Limerick.
During construction of the Limerick plant, our philosophy has always been to build quality.into'the plent rather than to attempt to " inspect in quality."
To that end, our quality program begins with the individual craftsman.
He is provided the requisite training, drawings and other work instructions to properly fabricate or
. install an item to its design.
The craftsman is directed in his efrorts by Bechtel Supervision.
Bechtel Construction Engineering provides the initial instructions to the crafts L
and supervision through work packages prepared for the performance of specific work.
Construction Engineering,
remains available while the work is being accomplished to answer any questions, resolve any technical problems, and provide additional project drawings / instructions for accomplishment of the work.
Construction Engineering'also inspects the work after completion, to ensure that the work meets specified design and installation requirements.
This inspection is conducted prior to turning the installation over to-Quality' Control for their final acceptance inspection.
Selected in process Quality Control inspection activities are also performed as part of the quality program.
The Bechtel Quality Control Program is fully described in the Bechtel Field Inspection Manual and Project Special Provisions.
It is a thorough and comprehensive program that defines the' certification of Quality Control Engineers (QCE),
the inspection process, nonconformance reporting systems, record retention requirements, and Quality Control involvement in the turnover process.
Prior to performing any work, the QCEs are qualified and i
certified in the discipline for which they have been j
assigned.
This qualification ensures their overall general knowledge in the particular discipline.
In addition, specific training is provided to ensure the QCEs are knowledgeable on specific project instructions and details.
J In order to provide the QCE with applicable inspection criteria, Project Quality Control Instructions (PQCIs) were developed.
The PQCIs identify the inspection criteria as well as specifying the inspection points.
The inspection criteria are obtained from project design specifications, i
drawings and instructions.
Prerequisites are identified to familiarize the inspectors with the details of the inspection.
In-process inspection points are identified.
These provide for surveillance inspections that are _- - _--_
9 Attachm2nt 1
+
accomplished as the' work progresses.
The surveillance inspections provide a level of confidence that the work continues to progress in a satisfactory manner.
The surveillance inspection reinforces the philosophy of building quality into~the product'.
If a problem develops with the work, it is our intent to catch the problem early and correct
- it before any significant repair or rework is required.
Ultimately, the PQCI provides.for final inspection of-the activities / components to assure that they. meet the design
- requirements.
The PQCIs were developed based on significant past' experience both at Limerick Unit 1 and other Bechtel projects and have been refined as this project progressed.
The PQCIs discussed were' prepared by Bechtel but received l
PECo review and/or approval.
i I
L Once equipment or a commodity system is accepted by QC, the j
inspection process does not end.
Two additional overview inspections are performed.
The first is part of the system turnover program.
Prior to system turnover to PECo Startup, a walkdown is performed by a multidiscipline. team of Construction Engineers, Quality l
Control Engineers and Startup personnel.
This'walkdown is l
performed to ensure that construction is complete and that any damage is noted.
When turned over to Startup, the system is tagged'to cle.arly identify that the system had been placed under PECo Startup control.:
The second inspection is made during the facility turnover process.
Again, a walkdown is performed by a multidiscipline team.
This team ensures that the facility and its components i
are damage free and again notes that construction in the defined area is complete.
These areas are then locked to the extent possible to prevent unauthorized entry.
Although this concludes the major involvement of Bechtel QC, l
it is not the end of the Limerick Quality Program.
In order to ensure the system functions as designed, ma extensive preoperational (Pre-op) testing program has been established.
This Pre-op testing is accomplished in accordance with j
approved procedures and by qualified personnel.
In order to ensure that the entire Quality Program is thorough and has functioned as established, audits are performed of all elements of the program by both Bechtel and PECo QA groups as appropriate.
These audit activities have found some problems, but in general, the Quality Program has been effective and produced the results desired.
In our response to COR-056, we noted that the quality program provides a graduated program of assurance that a level of quality is maintained on the safety related equipment sufficient to assure its function. l I
1
FT---
EE v..
1 The., intent of-the Limerick. Quality Program is to ensure that the installation meets the design requirements.
This is accomplished by building in the quality during the initial-installation..Then, an independent inspection is performed on'all quality related items to ensure the design is met.
In addition to the inspection at the time of installati'on, we have had both. system and facility walkdowns to-ensure.that unauthorized changes were not made'after the initial' acceptance.
During the preoperational phase of the plant, we ensure that the components and system function as designed, l
and have' program controls which assure design changes'are H
~ tested.
This is what was meant by the graduated program'to assure the. design function.
In summation, the quality program at: Limerick was established I
to ensure the highest level of quality that could be.
achieved..The plant is constructed to' ensure the design requirements are met and then the plant-is tested to ensure.
')
the design 1 function of the equipment is verified.
Since root cause analysis is not a precise science,4 root causes can be evaluated differently depending on a person's background
)
and experience.
We believe, that for the most part, we have established the correct root cause and.provided sufficient
.information to' properly address the concern.
However, we recognize thefinformation that'was supplied as part of the original response could be viewed in a few instances as requiring more detail or clarification as to methodology.
We appreciate the opportunity to amplify the information that j
was previously provided.
l 3
l s
i j
i
/ L______________
i
.. c= - -
?.
T e
l
%v
~
S -
1.
-Sectionc4'.l.10-?LPCI-Nozzle-to-safe'end welds (COR-007)-
Concern-During their construction' assessment, SWEC identified 1 radiographs:from three welds (LPCI-N17B-45,:LPCI-N17B-225,
'LPCI-N17B-315);thatfcontained discontinuities which were not documented ~or. evaluated ontthe radiographic. report'. form.
SWEC.also; identified one' discontinuity as;a root concavity in weld N17B-225.and found that':it did'not~ meet the, requirements of ASME'Section III, NB4424,~1974 Edition.
q Bechtel responded-.to'the root-concavity in weld N17B-225, but.
did not' reference, welds N17B-45.or N17B-315 in the. response-l tto'the COR.
Response
In response to.COR-007, we chose to address only'one'of'the four welds which Stone and' Webster' questioned. :WeTdid.this since each individual concern had been addressed as part of-CAI-022..Weinow recognize that it would-have been clearer to reference CAI-022 in_our response to COR-007.
COR-007 addressed LPCI nozzle-welds N17B-225, N17B-315,'and N17B-45.
It identified that certain indications were not initially evaluated.
The indications were addressed-in'our response'to CAI-022 and the-indications were found acceptable..The r,esponse was accepted.since Stone and Webster knew each area of the~ radiographs they questioned had been evaluated'under CAI-022.
i 2.
Section 4.2.1 - Reactor Building Unit Cooler Foundation Bolting (NRC Issue) l l
Concern This item concerned missing washers and one loose vibration j
pad mounting bolt on Unit Cooler 2FV211.
The licensee concurred with this finding and inspected the 20 unit coolers.
After identifying several additional discrepancies,
'l the licensee corrected the discrepant installations and-
)
trained QC and construction engineering personnel on the'need
'l for reviewing drawings and manuals in detail.
The team-considered these actions to be adequate to resolve the identified deficiencies.
l L
a 1
~
}3 -
[
I Attachmsnt 2
{
rg I
4wHowever,--theEteam found-thatLthe;11censee's technical 1 evaluationLof the significance of the loose mounting bolts wasinot. complete'infthat?the! response: stated only that'the Lexist'i'ng1 thread engagement was adequate'to transfer 1 design-tensile loads to the bolts. -This~ evaluation,did not address
- the; potential for:and ramifications'of total' disengagement-of the11oose fasteners from the vibrating unit.
Suchtl "h
disengagement would. render.the. fasteners unavailable;tol j
~
accept a load.
q 1
. Response The fanLmounting. bolts connect'the fan' foot.to the unit' cooler's base via'a'Fabreeka' absorption pad.
The 1/2"L
_ diameter bolt passes through a steel' washer, a/Fabreeka.
. washer,.the fan's support foot, a Fabreeka. pad,.the unit cooler's channel' base, and, finally,-l threads through a nut.
H AsLnoted previously by.NRC, the fanimounting-bolts lacked adequate 1 thread engagement-by 1/2 to 3: threads.
The~1acklof engagementJwas due to the bolt-lengthiand~not a. loose nut; theynutLwasisnug-tight against_the channel base.
TheLlack of full thread engagement raisesLtwo concerns to'be
' evaluated'to determine _ possibility.of' failure: a):could the 1
bolts' reduced tensile capacity.'have~affected the-unit cooler's safety function, and b)' could.the nut loosen..Since
.the threaded. portion' governs the bolt's. tensile' behavior,' the bolt'sireduced' tensile capacity due,to the reduced thread engagement"was evaluated.- The bolts with' reduced thread engagement adequately transferred the required. tensile: loads.
1 In addition'to.the tensile capacity requirements, snug-.
3 tightening the bolt induced small; clamping forces inLthe bolt that prevented movement 1of the connected members'in'the axial l-direction of the bolt, and prevented loosening of the' nut,
.(ref., " Guide to Design Criteria for Bolted and. Riveted' l
Joints," by G. L. Kulak et al.,.2nd edition 1987, pp.135).
Therefore, the fan's mounting condition'did not adversely impact the unit cooler s structural integrity or function.
j 1
l 1
l L3.
Section 4.2.2 - Unit Cooler Loose Brace Bolting (NRC Issue)
Concern This item concerned three loose fasteners connecting the fan suction bell and Unit Cooler 2FV210.
The licensee concurred l
with this finding and reinspected the safety related unit coolers, identifying one additional unit with similar discrepancies.
The licensee corrected the discrepant conditions and issued a field change request (FCR) to specify
7 mr.
, d i-
?
a.
3 +
Attachmant 2 stighteningTrequirements forLthese fasteners.
The1 team Reconsidered:these-corrective'actionsladequate,Jand no further action'is required of the. licensee.-
However,. the licensee's. response stated"that ;the !cause-of this ' discrepancy: wasL determined. to be a~ lack of ? tightening -
criteria-in vendor manuals and' drawings land that the original Installation.was suitably snug.1 The NRC team? found :that this statement could.not be supported and.that:the possibility-Texisted that the fasteners were not tightened or were improperly tightened'during initial installation'. :The
-licensee-didinote in the response that action to. preven't' 1
recurrence: included the_ retraining 1of personnel:and issuance
.of a corrective action = report against the QC-inspector who.
- had performed the original acceptance-inspection of theseL coolers.
However, the-licensee,'s failure to acknowledge specifically that the cause of this discrepancy.could-have been personnel' error.is an example of an uncompleted statement of root cause..
Response
I.
~Upon further evaluation of this condition and the' causal.
' factors 1 presented in response to' Inspection Report 88-202, we agree'that personnel error contributed-to.this. deficiency in conjunction with'the lack of instruction' contained in.the vendor manual / drawing.. The: Field Change request, ME-1419, which was issued to clarify the tightening requirements.
As a' result, Construction Mechanical Engineering; conducted 1 training emphasizing thorough drawing / manual review prior to field-installation.
The. deficiency identified with these bolts on unit-cooler 2FV210 should have been identified via a Field Change: Request by Construction Mechanical Engineering '
prior to installation.
l i
4.
Section 4.2.3 - Loose Bolts on HVAC Duct Segments (Nhc Issue) i Concern 1
This item concerned two loose bolts joining HVAC duct i
segments.
The licensee concurred with the finding, tightened I
the bolts, and inspected 92 additional joints.
The team l
considered the corrective actions adequate, and no additional action is required of the licensee.
j b
I L
L !
1
Attachmnnt 2 However, the. licensee attributed the cause of the loose bolts to unauthorized alteration of the installation after final QC inspection.
This statement of cause was too restrictive in that the licensee did not acknowledge that inadequate i
installation and inspection was an equally possible cause.
Response
i We acknowledge that inadequate installation and inspection represents a potential cause for this deficiency however, we f
disagree that these contributing factors are necessarily equally possible causes.
Based on our re-evaluation of the condition, its age, extent and possible causal factors we have concluded that unauthorized alteration after final QC l
inspection is the most likely cause of this deficiency.
5.
Section 4.2.4 - Improper Torque on RHR Pump Disc.large Head Bolts (COR-016) l Concern SWEC found residual torques on the RHR pump discharge head i
bolts that were less than the vendor specified minimum installation torque.
The licensee retorqued these fasteners on all RHR pumps and on the core spray pumps, which are of the same design.
Bechtel identified torque values that were below and above the specified values.
The licensee l
established a periodic program to torque the affected bolts I
and monitor the torque values obtained to better assess this phenomenon.
i The NRC team noted that the Bechtel response identified the
?
probable cause of the low torque values identified by SWEC as gasket creep with other contributing factors that would cause the loosening of bolts.
However, the Bechtel response to the COR failed to address the fact that some residual torque values obtained during the follow-up inspections exceeded the installation torque, although the information was available in the documents accompanying the COR response.
The excessive torque values would not be consistent with gasket creep or the contributing factors noted as possible causes of the out-of-tolerance torque values.
This is an example of an incomplete statement of root cause.
Response
The RHR and core spray pump discharge head bolts were torqued between 3/2/89 and 3/6/89 beginning the surveillance cycle.
l j
)
M,'
- .g M,
A j.
41Retorquing h'as~been. conducted.on three additional' occasions.
(4-1-89/4-2-89, 5-4-89 and 6-2-89).
One full quadrant of each RHREand Core Spray Pump bolts (15'each and 7 each-m:
_respectively) were scribed"andimonitored.
LThe~ procedure utilized to. verify boltatorque,2 consisted 1of'
~
inspectingithe" scribed lines'priorito torque' verification, verify; torque,: inspect scribed lines'after. torque verification,;and document'all; findings.
4 During torque ~ verification, 4 bolts on 2AP206 and'27 bolts on.
2DP206 were found under' torqued..The as-found. torque values were recordediandLthe. bolts retorqued.to 475 ft/lbs.
The
' scribed quadrants'were checked for movement.-
The' actual ~
movement after retorquing was recorded.
The lines were rescribed on the bolts which moved (as applicable) to.
facilitate' continued observation.
To date,'no' movement has been noted prior to torque
. verification.
On~each.of these subsequent reverification out.of tolerance torque values were discovered, and corrected, sand the'results-were, forwarded to General Electric for their evaluation.
Evaluation of'the root cause and other contributing factors
- is continuing to. adequately; identify a permant solution.
.When.the' initial' response.was prepared'for this concern, we had only collected data from one torque verification _and as a result of'that data we formulated'our response'to the-concern
-based on what we felt.to be'the most logical: set of
' circumstances ' to identify this ; problem.
.15. 'Section 4.2.5 - Incorrect HVAC Hanger Material Size (COR-019)
Concern This item concerned the installation of an incorrectly sized-gusset plate on'HVAC hanger 66.
Specifically, the installed gusset plate was 1/2-inch thick, as specified on drawing M-1169-2.
According to an FCR written to change the drawing and listed in inspection documents as the applicable design change document, the correct gusset plate thickness was 3/4-inch.
Bechtel concurred with the observation, performed calculations that indicated the existing hanger configuration was technically adequate, and revised the affected drawings.
The team considered this assessment and corrective action to be adequate, and no additional action is required of the licensee.
l l _ _ _ _ _ _ _ _ _ _ _ _ _
1
H l
1 I
.l 1
Attcchm nt 2 j
i However,'Bechtel's' response to the COR indicated that this L
discrepancy resulted from a mistake by the construction 1
engineer in preparing the FCR.
Bechtel stated that, to determine the extent of this discrepancy, six similar hanger installations and 15 FCRs prepared by that construction
-engineer were reviewed, and no further discrepancies were noted.
The response also indicated that the action to prevent recurrence was to emphasize to the responsible construction engineer the need to pay more attention to detail.
Although the field engineer may have erred in specifying the plate size on the FCR, Bechtel's response did not acknowledge that the support was not constructed in i
accordance with the specified design and was accepted in error by the QC inspector.
Because Bechtel did not properly identify inadequate installation and inspection as a cause of this discrepancy, its corrective actions were not properly focused.
However, due to the minor nature of the installation discrepancy and lack of similar deficiencies identified by SWEC and the NRC, no further action is required of the licensee.
Response
1 We have reviewed the causal factor in the original response and recognize that it is confusing.
The original design called for a 1/2 inch plate in accordance with Drawing C-i 1399.
Due to a constructability problem, Field Change Report (FCR) C-ll860F was generated which changed the plate size to 3/4 inch; however, this change was not noticed and a 1/2 inch plate was installed contrary to the FCR requirements.
The j
cited condition is attributable to an oversight on the part of Construction Engineering and Quality Control.
Both the l
Construction and Quality Control Engineers were aware of the
[
original design requirement and incorrectly inspected the hardware to the original requirements rather than the revised requirements.
An FCR was not used to document an after the fact nonconforming condition, as can be inferred from the original causal factor.
Retraining was performed for both i
groups to ensure they take appropriate care during the preparation of the FCR and during the inspection process.
7.
Section 4.2.6 - Lack of Installation and Inspection Criteria for RHR Heat Exchanger Lower Supports (NRC Issue)
Concern The design specification and the vendor technical manual for the RHR heat exchangers indicated that installation requirements should include clearance requirements for the _ _ _ _ _ _
~
q, "
I
)
Elowgr-support:boltsito-allownfor, thermal-expansion.
The, team
- wasuconcerned that the installation-' documents'for'the RHR heat exchangers did notispecify any required clearance,.and; Lthus,'the existence or' amount'of any clearance'in the as-
~,
. built-condition was unknown. 'Although;the111censee stated in
'the: response that it did not concurtwith1this concern,.the-licensee conducted inspections'to verify'the radial
. clearances'for the RER-heat exchanger lowerusupports and performed calculations to determine that thermal: growth would not"cause unacceptable' loads to the heat exchangerror 1
supports.
The. team still considered this a failure to incorporate design criteria into' installation documents.
In fact, the.
licensee's subsequent inspections and reviews also identified the existing clearance between the support feet ~and' adjacent bumpers as being outside the. range used in the design.
analysis.' The' team considered this.another example of-improper incorporation of design criteria.: Bechtel performed stress 1 calculations using' combined seismic and thermal loads and.found that FSAR-allowable values would be exceeded, although an analysis indicated the support steel would remain functional.
.To' comply with the FSAR commitments,.the licensee. installed shims where required.on the lower support bumpers'..
InJaddition, the NRC inspection. team noted that the licensee's=re3ponse to'IR 50-353/88-202 did not indicate-that-FSAR commitments would have been exceeded lwithout' installation ~of the shims.
Although this is an example of an-H incomplete response,.the' licensee's corrective actions were-adequate,:and no additional action is required.-
Response
'In our response to this. issue, we stated that the shims were addedLto limit the gap between the heat exchanger foot'and lateral bumpers to satisfy the original design, calculations.
1whichfare the basis for the FSAR. allowable stresses.
We agree that the original response should have explicitly stated that the shims were.added to ensure compliance to the j
FSAR' committed stress allowables.
1 h'
8.
Section 4.2.7 - Missing Jam Nuts on Equipment Foundation Bolts (NRC Issue)
Concern This concern related to the licensee's failure to install jam nuts on foundation bolts for various pumps and unit coolers in the reactor building, as required by the general -
1 i
foundation drawing. The licensee concurred with this finding and performed an inspection of all mechanical equipment foundations in the reactor and diesel generator buildings.
After identifying additional cases of required jam nuts and washers that had not been installed, the licensee corrected j
the discrepant conditions or deeraed them as acceptable on a case-by-case basis.
The team considered the licensee's i
actions to be adequete.
]
i However, the licensee indicated in the response that the
]
basic cause of these discrepancies were inexplicit and
]
inconsistent equipment foundation drawing details.
The last sentence of the paragraph indicated that, "In addition, an adequate drawing review was not performed in all cases.
The team noted that the same drawing requiring jam nuts applied to all of these foundations.
The licensee's response directed attention to poor documents, rather than to the fact that the installers and inspectors failed to comply with i
existing, referenced design drawings.
Although no additional action is required of the licensee, this response was another example of an incomplete root cause determination.
Response
)
i In our response to the NRC Inspection Report, we noted two causes for this problem.
One being that the drawings were not consistent, and the other that an inadequate review was performed by the Construction and Quality Control Engineers.
We did provide more details on the problems associated with the drawings, and as a result of this level of detail, it could easily be inferred that we considered this cause to be l
the more significant problem.
This was not our intent.
The notes on the Civil Drawings required the use of washers and jam nuts.
We concur with your assessment that the overriding cause was the inadequate drawing review by Construction and QC personnel.
In fact, Construction and Quality Control personnel were trained on the need to properly research the applicable drawing.
9.
Section 4.2.8 - Discrepancies in the Ccntrol of Work Packages (NRC Issue)
Concern During its inspection, the NRC team noted numerous discrepancies in the documentation related to work packages for mechanical equipment installation and modification.
The team was also concerned with the complexity and clarity of inspection documents and considered these factors possible contributors to some of the discrepancies noted on mechanical L____
)
N'
<*l s.
. c l
equipment installations.- The 11censee stated in its' response.
to this issue that~it did not concur with this concern, indicating that the work packages are neither quality
]
documents nor permanent records.
The response.further stated
'j that all QC-required data and inspection attributes had been accounted for, and numerous audits had determined'these programs to_be adequate.
The team identified this concern to the licensee as an observation of apparent program weaknesses that have also been identified at other nuclear plants-where'they have directly contributed to installation deficiencies.
The team I
reviewed two Bechtel audits and nine PECo audits provided by the licensee in support of the response.
The team noted that discrepancies similar to those addressed in the NRC report, such as missing or improperly referenced acceptance. criteria and other inspection and installation documents and improper inclusion of installation details in work packages, were identified in four of the audit reports provided by the licensee.
Several cases of_ performance and OC acceptance of out-of-tolerance shaft alignments were also identified.
In addition, the team noted that the poor quality of the recently issued Modification / Rework _ Packages (MRP) for the core spray pump discharge head bolting program, discussed in Section 4.2.4 above, was a current example of this concern.
Although no additional action is required of the licensee, the licensee indicated following the inspection that they recognized the NRC's concern regarding adequately documented work packages.
Response
In order to ensure that document control problems with work packages would not lead to hardware problems, we instituted several control mechanisms.
Not all drawings were put in the work packages.
Those required to be only referenced were placed on controlled field sticks and were available for reference.
In addition, we did not allow the Quality Control L
inspections to be performed using the documents in the work
[
package.
The Quality Control Engineers independently i
determined the drawings required and their revision status.
Because of the number of packages involved, Bechtel Construction Engineering performed self audits to ensure proper document control.
In addition, both Bechtel and PECo g
E Quality Assurance also audited work packages as a document control function and as part of the hardware audits.
l I
l,
l i
- a