ML19325D970
| ML19325D970 | |
| Person / Time | |
|---|---|
| Issue date: | 09/29/1989 |
| From: | Potapovs U, Wilson R Office of Nuclear Reactor Regulation |
| To: | |
| Shared Package | |
| ML19325D931 | List: |
| References | |
| REF-QA-99901164 99901164-89-01, 99901164-89-1, NUDOCS 8910270242 | |
| Download: ML19325D970 (10) | |
Text
_.
e ORGANIZATION: WHITTAKER CORPORATION SIMI VALLEY, CALIFORNIA REPORT INSPECTION INSPECTION NO.: 99901164/89-01 DATE: June 20-22, 1989 ON-SITE HOURS: 22 CORRESPONDENCE ADDRESS: Mr. Tincthy E. Conver, President Electronic Resources Division Whittaker Corporation 1955 N. Surveyor Avenue Simi Valley, California 93063 ORGANIZATIONAL CONTACT: Mr. Fred P. Rudek, Manager Quality TELEPHONE NUMBER:
(805) 584-4100 NUCLEAR' INDUSTRY ACTIVITY: Manufactures metal-sheathed cable and connector assemblies for numerous Combustion Engineering (CE) core exit thermocouple and reacter vessel level monitoring systems, and for General Electric in-core nuclear instruments.
Sales are divided roughly equally between nuclear and aerospace applications, although the nuclear application was developed only within the last ten years.
ASSIGNED INSPECTOR:
2 h
R. C. Wilson, Senior Reactor Engineer Date~
Reactive Inspection Section No. 2 (RIS-2),VendorInspectionBranch OTHERINSPECTOR(S):
j APPROVED BY:
bd
$h_P
'PM%
Uldis Potapovs, Chief ( Reactive Inspection Section No. 2 Date (RIS-2),VendorInspectionBranch INSPECTION BASES AND SCOPE:
A.
BASES:
10 CFR Parts 21 and 50 B.
SCOPE: This inspection was made as a result of an allegation concerning lack of adequate quality assurance at Whittaker, including improper use of silicone material in place of hermetic seals, in safety-related cable and connector assemblies for nuclear power plants.
PLANT SITE APPLICABILITY: Numerous o
99901164
. ~.
e
~
~
y
.3 i
g-ORGANIZATION: WH1TTAKER CORPORATION
)
m:.
SIMI VALLEY. CALIFORNIA q
REPORT 1
INSPECTION l
NO.: 99901164/89-01 RESULTS:
PAGE 2 of 10-2 j
A.
VIOLATIONS:
None-
)
B.' NONCONFORMANCES:
None C..
UNRESOLVED ITEMS:
H None O.
STATUS OF PREVIOUS INSPECTION FINDINGS:
None. This.was the first NRC inspection of this facility.
'E.
INSPECTION FINDINGS AND OTHER COMMENTS:
1.
Allegation Scope The allegation consists of two parts, both related to the manufacture of cable and connector assenblies supplied for safety-related applications in nuclear power plants. The concerns are as follows:
a.
It was alleged that a silicone sealant was used improperly and without the knowledge of. customers, principally Combustion Engineering, Incorporated (CE), prior to Septenber 1988.
b.
It was alleged that Quality Assurance (QA) for the manufacture of the assemblies was inadequate.
These two concerns were conveyed to Whittaker and were dddressed during this inspection.
To obtain relevant information concerning customer requirements, a previous inspection was conducted at CE on May 22-25, 1989 as documented in Inspection Report 999000401/89-01.
2.
Use of Silicone Sealant by Whittaker The inspector initially understood this allegation concern to relate to the cable end seal discussed in section E.1 of I-Inspection report 99900401/89-01.
Information provided by Whittaker redirected the concern to address a silicone 1
L i
i,
~
r F
t-
'.b n.
' ORGANIZATION: WHITTAKER CORPORATION L-SIMI VALLEY, CAllFORNIA l
REPORT INSPECTION NO.: 99901164/89-01 RESULTS:
PAGE 3 of 10 coating applied to the ceramic bead inserts in the connector design manufactured from 1984 to 1988.
i The product under discussion consists primarily of ceramic.
insulated, netal-jacketed multi-conductor electrical cables with connectors at both ends.
Further description including applications is addressed in the inspectior report cited above.
Prior to 1984, CE provided Whittaker with connectors from another manufacturer for assenbly to the Whittaker cables.
In 1988, Whittaker began manufacturing a new connector design which does not use the silicone sealant. The design introduced in 1984 included a metal header plate with holes for the electrical contacts.
Each contact pin was surrounded by a small ceramic bead.
The assembly was oven-fired to fuse the ceramic and produce ceramic to metal hermetic seals. Depending on tolerances on the pin and hole diameters, header thickness, and quantity of ceramic, the molten ceramic could slightly under-or over-fill the annular gaps. The resultant meniscus could result in cracks in' the surface glaze near its edge. The cracks could affect both leak tightness and electrical insulation resistance. The deficiency was detected by an insulation resistance (IR) test on the connector prior to its assembly to cable.
The inspector reviewed production control route cards for 181 connectors to assess IR test failure rates.
Interviews with experienced production, manufacturing, quality, and engineering personnel did not produce a consensus regarding the percentage of connectors that f ailed the IR test or any trend in that percentage.
Thc personnel interviewed stoted that for connectors that failed the IR test, a silicone fluid coating was applied to the insulators, followed by an oven baking. The personnel i.
stated that 11 the connector passed the second IR test it L
was accepted; if not it was scrapped.
For the production control route cards reviewed by the inspector, all of the silicone treated connectors (slightly less than 10 percent of the total) passed the second IR test. The treatment process is straightforward. One nianager stated that although not l
called out in procedures, it was considered a " standard rework,"
L but its use was subject to approval by a manufacturing engineer.
l l
ORGANIZATION: WHITTAKER CORPORATION SIMI VALLEY, CALIFORNIA l
REPORT.
INSPECTION NO : 99901164/89-01 RESULTS:
PAGE 4 of 10 The silicone fluid used by Whittaker is described in the-manufacturer's product sheet as a water-repellant coating.
Whittaker has used such products for 20 years, as evidenced by a 1969 manufacturing process specification for a similar coating, written to meet a Navy department specification.
The same coating is covered by a 1982 Whittaker manufacturing process specification for treating machined ceramic parts used 6s filler blocks in the assemblies provided to CE; in that ap lication the machining removes surface glaze, and the si icone treatment is used for cleaning and to provide a final surface conditioning. The silicone fluid and solvent addressed by the 1982 specification is the same as was subsequently used on the multiple-bead connector headers.
Manufacturing procedures did not formally address the silicone coating for connector headers until April 20, 1987 when a new manufact.uring process specification was issued.
Prior to that time procedures did not address the header coating, nor did they address whether or why it would be applied. The only documentation of its use appears to be hand-written addi-tions on the otherwise typed production control route cards.
Whittaker management personnel explained that the division strongly relied on proprietary protection for its products for many years. Both military and civilian customers have agreed that the customer controls the product down to the top assembly level, and Whittaker controls below that level.
Information provided to customers normally consisted of the following:
top level assembly drawing, weld schedules and samples, acceptance test procedures and data, and rework i
procedures.
(This statement appears to be consistent with files reviewed by the inspector at CE.)
Preparation of the 1987 coating specification appears to be an early step in improving internal process controls at Whitta ker. The May 1988 CE audit clearly added impetus to that effort. Whittaker provided the inspector with a copy of a September 1988 specification control diawing that was submitted to and approved by an aerospace customer covering silicone treatment of connector header seals; this was cited as an example of a policy change toward providing more information to customers.
The silicone treatment is not employed with the new single bead header connector design now supplied to CE.
.~.--.......
.... ~. -.. ~...
n 1
f.
Am ORGANIZATION: WHITTAKER CORPORATION-SIMI VALLEY, CALIFORNIA i
REPORT INSPECTION N0.: 99901164/89-01 RESULTS:
PAGE 5 of 10-
'With respect to 10 CFR 50.49 environmental qualification of theoriginal(1984-87) Whittaker connector design, CE's l
qualification basis was addressed in an earlier Inspection Report 99900401/88-01 dated January 10, 1989. The qualification basis appears acceptable subject to the question of whether the qualification test specimens had received the silicone treatment. Whittaker personnel were unable to locate appropriate records; the search is conplicated by a 1987 relocation.
There is no basis for con::1uding that the test specimen connector headers had received the treatment, although it is possible, based on present knowledge. The possibility that treated connectors may represent an EQ concern does appear remote, for the following reasons:
a.
Both CE qualification test programs, as well as Whittaker's recent testing of cable / connector assemblies for boiling water reactors, included machined ceramic blocks that were treated, and there is no evidence of poor electrical perforraance by those blocks.
b.
The headers form only secondary seals in mated connectors, as would be of interest during plant operation.
c.
Protective caps with 0-rings and lanyards are provided to discourage moisture ingress when the connectors are not mated.
d.
The silicone dioxide cable insulation is relatively insensitive to humidity, unlike other ceramic cable insulation. Whittaker personnel stdted that it does not hydrolyze, and loses only about i te one decade of insulation resistance per year in ambient humidity if the cable end is completely open.
e.
The multiple bead connectors are no longer being manufactured for nuclear use.
Another seal concern raised by the alleger involves repair of connectors by drilling, then plugging, holes in their sides.
The inspector believes that this concern is the same as a problem cited in Inspection Report 99900401/89-01 at the end of section E.1, and the two will be addressed jointly.
(
l l
3 k
'0RGANIZATION: WHITTAKER CORPORATION SIMI VALLEY. CA'.lFORNIA 1
REPORT INSPECTION
.NO.:: 99901164/89-01 RESULTS:
PAGE 6 of 10 A November 13, 1987 CE. letter to Whittaker on Purchase Order (PO) i 9770018-71186 refers to five connectors supplied without a crimp sleeve for the ground contact, unlike 19 others that were specified as identical, all under CE PO 9770018 71186. Whitta ker contract records show that most of the connectors were manufactured j
and certified by buth Whittaker and CE prior to the initial submittal of the drawing to CE. Some significant. dates are as follows:
March 5,1987 - Whittaker internal release of drawing 16-99-008000 June 4,1987 - Whittaker test and inspection records and C of C 4
June 11, 1987 - CE Certificate of Equipment signed by engineering
^
and quality control " based upon source surveillance being performed at the vendor's facility" June 15, 1987 - First submittal to CE of drawing 16-99-00800 Following conditional CE approval, Revision A was made and was approved by CE on July 9,1987.
Of the 24 connectors shipped, 19 were built to Revision 0 and 5 to kevision A; the 2 differ in ground connection method.
Upon discovery of the different connector types by the third party to whom they were shipped, Whittaker prepared a Rework Procedure dated February 11, 1988 which was approved.via a Technical Change Request to CE. The rework procedure was reviewed and discussed with the Whittaker manufacturing engineer who supervised connector assen.bly at the time of rework.
It involved drilling an access hole through the side of the connector body, inserting a solder cup and threaded stainless steel rod, replacing potting material, and welding and blending the rod to the connector body.
The potting material performed no sealing function; it filled void space to support wires.
The reworked connectors appear to be satisfactory. However, it is clear that Whittaker mhnufactured and shipped connectors to an unapproved or6 wing, and CE approveo the shipment; evidently neither acted further until they were informed by a third party that two different types of connectors were delivered.
Summarizing the silicone treatment concern, the NRC inspector concludes the following:
a.
An unknown quantity of connecturs, possibly dozens, was supplied by Whittaker to CE with a silicone fluid treatment r~-,-n--,,-~<.-.,---.m,-e--e,..
e em-,-
,,,,,-w-,,-,a
-,--r-.-
w-,---
ORGANIZATION: WHITTAKER CORPORATION SIMI VALLEY, CALIFORNIA
?
REPORT INSPECTION
- NO.:
99901164/89-01 RESULTS:
PAGE 7 of 10 4
applied to the connector header insulators. Written procedures did not address when or how to apply the treatment until April 1987. The treatment improved performance with respect to two procurement specification parameters, leakage and electrical insulation resistance.
4 b.
The treatment was not used on the relatively saml1 number of nuclear safety-related connectors supplied to General Electric Company, and is no longer used in the manufacture of any nuclear safety-related connectors.
c.
The treatment is not believed to constitute a nuclear safety concern for reasons stated above.
d.
Information. gathered during investigation of the silicone sealant concern clearly demonstrates that both Whittaker and CE f ailed to satisfy Quality Assurance concerns related to this equipment. This subject is addressed in the next section of the inspection report.
3.
Adequacy of Whittaker Quality Assurance In detailing this concern, the alleger stated that during the first nine months of 1988 certain process specifications and workmanship standards were not available on the production floor, and that manufacture was accomplished to drawings handmarked by engineering to reflect customer order information. The NRC inspector determined at CE.(see inspection report 99900401/89-01) that CE audits of of Whittaker in May and September of 1988 had revealed similar concerns, and CE and Whittaker had initiated actions to address the concerns.
Whittaker quality assurance personnel told the NRC inspector that some components were built to drawings marked and signed by the project engineer, but that final inspection was performed only to approved drawings, and items could not leave the area without an approved Engineering Change Order.
Whitta ker management personnel also stated to the NRC inspector that documentation had been "not truly acequate," and that documentation was " poor." Reliance was ultimately placed on production personnel to ensure that nenufacturing operations were correctly performed.
In turn, experienced assembly personnel stated that any deviation had to be approved by someone not on the production floor (engineering, manufacturing engineering, or project engineering). Assembly personnel also commented that they sometimes made notes to themselves on drawings.
4
3 Le.
i ORGANIZATION: WillTTAKER CORPORATION SIMI VALLEY, CAllFORNIA l
REPORT INSPECTION NO.: 99901164/89-01 RESULTS:
PAGE 8 of 10 l
Both the alleger and the 1988 CE audits noted the.t sales work orders were used to impart purchase order requirements to
{
production, serving as the final inspection checklist.
Whittaker relocated from Burbank to Simi Valley, California --
about 30 miles -- in the spring of 1987. The move complicated access to some records.
For example, during the NRC inspection a box of production route cards covering early 1987 could not be located, although adequate alternate information was provided to l
the inspector. The relocation was obviously another complicating factor in records control.
i Whittaker management stated that the company has always been i
straightforward with customers in infoming them that the j
basic processes are Whittaker's and are not subject to customer i
controls.
In the case of the silicone fluid treatment, apparently CE was not made aware of the existence of the process l
until af ter the new single bead header connector design n.ade the i
process no longer applicable to CE contracts.
All of the CE P0s reviewed by the inspector invoke Revision D of CE Specification 00000-WQC-11.1, " Supplier Quality Controi Program Specification for Quality Class 1 Equipment or Services" dated October 4, 1974. This specification for Quality Class 1 equipment covers much of the scope of Appendix 0, but it provides CE with options concerning the extent of oversight applied to Whittaker.
Insofar as CE did not review the processes and methods by which Whitteker manufactured equipment, deficiencies in those processes and cethods were not identified by CE.
i Even though CE may have elected not to address manufacturing processes, at least until 1987 or 1988, it is nonetheless clear l
that Whittaker's failure to have a procedure covering the i
silicone fluid treatment constitutes a nonconformance tu section 6.5.1 to of the CE quality specification, which requires that activities af fecting quality must be controlled by written procedure.
Shipment of the connectors under P0 9770018-71186 without drawing approval is a nonconformance to Section 4.1.1.1 of the CE quality specification, which requires a vendor program that assures quality throughout all areas of contract performance.
'.his is an anomaly f or which Whittaker had no explanation.
I l'
1
- 6 L
i ORGANIZATION: WHITTAKER CORPORATION SIMI VALLEY, CALIFORNIA l
I REPORT INSPECTION NO.: 99901164/89-01 RESULTS:
PAGE 9 of 10
)
i s
i The ongoing activities by Whittaker and CE appear to be correcting the deficiencies in Whittaker's quality program.
c Several changes were noted as examples of quality improvenents:
a revised set of controlled manufacturing procedures for connectors; computerized standard routing lists with specific j
references to applicable procedures; computerized sales order descriptions providing all department with summaries of contract requirements; the Quality Assurance manual revisions being reviewed by CE; and reorganization so that the Quality Assurance Manager now reports to the division president.
l i
Surmarizing the concern with respect to the adequacy of Whittaker's l
Quality Assurance, the NRC inspector concludes that deficiencies existed, most significantly with respect to the following:
a.
Lack of a procedure cuyering silicone fluid treatment of connector headers prior to April 1987, b.
Shipment of connectors under PO 9770018 71186 without drawing approval in 1987.
These deficiencies are primarily historical in nature, and are considered to not have resulted in shipment. nf equipment that constitutes a nuclear safety concern.
Numer;as improvements in quality assurance have been noted since the CE audit in May 1988 i
and the NRC will continue to monitor Whittaker's and CE's actions in this regard.
f
[
_ _. _ - _ _ _ _ - ~.
I ORGANIZATION: WHITTAKER CORPORATION SIMI VALLEY, CAllFORNIA REPORT INSPECTION NO.: 99901104/89-01 RESULTS:
PAGE 10 of 10 j
i F.
PER$0flS CONTACTED:
- +T. E. Conver, President
- +R. H. Mauldin, Vice President, Engineering i
- +R Snyctr, Vice President Customer Service Operations
- +T. P. Rucek, Manager, Quality l
- +H. E. Simpson, Program Manager
- +J. Sobelman, Manager, Manuf acturing Engineering l
- +T. 5. Whitehead, Manager Manufacturing l
- +G. P. Casey, Manager, Finance; Contract Administration
- +R. Lolley, Manager, Field Service and Training
- +0. MacLean, Manager, Materials
+J. Torres Supervisor, RF and HJTC Manufacture
- J. Thomsen, Supervisor, CA Test
+J. Denny, Hanuf acturing Engineer, Hermetics l
+R. Cortez, QA Inspection D. Carle, QA Inspection
- 1. Reinhart, Connector Assembly
- +C Reinhart, Project Engineer
- +A. Harootion, Project Engineer
- +K. Wilson, Engineering Tech Staff
- +R. Couser, Senior QA Engineer
+F. Roy, Engineer (Patel Engineering, Contractor)
- otttoded entrance meeting i
+ ettended exit meeting
)
t
[
i
=
e ORGANIZAT!0N7 COMBUSTION ENGINEERING, INC.
f POWER SYSTEMS GROUP WiknthD_ rol'L'rrTffff7 REPORT INSPECTION
\\
INSPECTION I
NO.:
99900401/89 01 DATE: May 22-25.1989 ON-SITE HOUR $t 24 l
CORRESPONDENCE ADDRESS: Mr. Walter D. Mawhinney, Vice President 1
Nuclear Quality Systems Cosbustion Engineering, Inc.
Power Systems Group 1000 Prospect Hill Road i
Windsor 06095 ORGANIZATIONAL CONTACT: StevenI. Connecticut r
TELEPHONE NUMBER:
(203) 285-5213Toelle, Manager, Operating Reactor Licensing l
j NUCLEAR INDUSTRY ACTIVITY: Co.tbustion Engineering, Inc. (CE) has had NSSS contracts for 16 domestic reactors, and has support service contracts for approximately 40 reactors worldwide.
ASSIGNED INSPECTOR:
x 9 41!,9 R. G. Wilson, Senior Reactor Engineer, Reactive InspectionSectionNo.2,(RIS-2) Vendor ae i
Inspection Branch OTHERINSPECTOR(S):
/
APPROVED BY:
AW U. Potapovs, Ghfef, RIS 2, Vendor Inspection Branch
' 9-f*I$
E 1.
l i
INSPECTION BASES AND SCOPE:
i i
A.
BASES:
10 CFR Part 21 and Part 50.
l B.
SCOPE:
1.
Review engineering and procurement records for thermocouple cable seats for safety related applications.
2.
Review 10 CFR Part 21 notification, dated April 20, 1989, concerning control element assembly slippage.
PLANT SITE APPLICABILITY: PaloVerde1,2,and3(docket 50-528,529,530);
Turkey Point 3 and 4 (50-250, 251); South Texas 1 and 2 (50-498, 499); Salem 1and2(50-272,311); Indian Point 2(50-247);WNP-3(50-508).
I I.
y i q m w[O V8 r
-,,.,n--,-,-n-
.-m
,-o,---,---+-m-,--we,-o-.,_m,.,---w,-.,mnnn,,----,--,,--.--m a,
,m,---e-o..e.-
e
--,,----,,-w-,,w,-,.-w,,
CRGANIZATION: COMBU$ TION ENGINEERING. INC.
s WikD50R, CONNECTICUT
~
REPORT INSPECTION NO.: 99900401/89 01 RESULTS:
PAGE 2 of 6 i
l
,I A.
_V!0LATIONS:
l None 8.
NONCONFORMANCES:
.i None I
C.
UNRESOLVED ITEMS:
.l t
Control of procurement activities, see report para 9raph E.1 l
I (8901-01).
D.
STATUS OF PREVIOUS INSIECTION FINDINGS:
Not applicable.
E.
INSPECTION FINDING $ AND OTHER COMMENTS:
i',
1.
Thermocouple cable end seals s
i The first topic addressed in this inspection was hermetic seals I
for cable assemblies used in two safety-related systems supplied byCE,coreexitthermocouple(CET)andreactorvessellevel i
metal jac)eted, mineral insulated cable with threaded electricalB I
monitorin!
(RVLM) syster.s.
t connectors.
Electronic Resources Division of Whittaker Corporation.The c Either the cable ends or the cable-connector junctions must be
't hermetically sealed to prevent moisture contact with the cable's i
silicon dioxide ceramic insulation and resultant degradation of i
insulation resistance. Procurenent criteria including s>ecifi-following P0s from, CE to Whittaker were(PO) req [uiren cations, drawin9s and purchase order or the i
reviewed:
Plant P0 Date Palo Verde 1 9170287 14273 03/26/81 i
Palo Verde 2 9170288 14373 03/26/81 Palo Verde 3 9170289-14473 03/26/61 i
Palo Verde 1 9271857-14273 09/10/82 Palo Verde 2 9271858-14373 09/10/82 Palo Verde 3 9271859-14473 09/10/82
i ORGANIZATION: COMBUSTION ENGINEERING. INC.
WINDSOR, CONNECTICUT REPORT INSPECTION i
N0.: 99900401/89-01 RESULTSt PAGE 3 of 6 l
Plant P0 Date 1
I Turkey Point 3 9270855 16081 04/28/82
)
Turkey Point 4 9270854-16081 04/28/82 Turkey Point 4 9270282-16081 02/13/84
{
i South Texas 1 9471519-4884 08/27/84 i
South Texas 2 9471520 4884 08/27/84 i
Salem 1 9770018-71186
' 01/15/87 Salem 2 9770019 71186 01/15/87 l
Indian Point 2 9874769 D9421 05/23/88 l
CE engineering personnel were interviewed concerning both i
i generic matters and these specific procurements.
Each CE PO invoked 10 CFR Part 21. Each order also applied CE's " Quality Assurance Requirements for Suppliers of Nuclear Safety-Related l
(CE Quality Class 1) 1tems and Services," Specification WQC-11.1 Revision D, dated October 4,1974, Quality Class 1 was l
specified.
Each P0 also applied specification 00000-FEA-6102,
- Engineering Specification for in-Core Instrumentation Intercon-nectin9 Cable Assemblies," Revision 04, dated January 15 1982.
Each PD also invoked a CE assenbly drawing providing plan,t-specific infortaation such as quhntity and length of cable assemblies, i
connector suppliers and types, and connector pin assignments, t
1 Originally CE supplied all of the connectors to Whittaker, as specified in paragraph 2.3.1 of Specification 6102 Revision 4 No credit was taken for connector hermetic seal capabilities.
i Paragraph 4.2.5 of Specification 6102, Revision 4 specifies that the ends of each cable shall be hermetically sealed to prevent i
moisture absorption, prior to connector attachment.
In 1984 CE docunented environmental qualification of theraccouple cable asseriblies using Whittaker-manufactured connectors; see NRC Inspection Report No. 99900401/88-01, dated January 1,1989 for discussion. The new Whittaker connectors replaced Litton-Yeam connectors in most instances. GLH Technology Corporation connectors continue to be used for the RVLM disconnect nearest the reactor and in some cases Swagelok connectors are used for connection to CET sensors.(not assembled by Whittak[tr)
CE personnel explained that connectors manufactured by Whittaker provide an adequate hersetic seal for the cable, and a separate cable end seal is not l
required.
t
_,...,.m..,_.
b ORGANIZATION: COM6USTION ENGINEERING, INC.
f WIND $0R, CONNECTICUT I
t REPORT INSPECTION i
NO.: 99900401/89-01 RESULTS:
PAGE 4 of 6 l
Typical recent Whittaker Acceptance Test Proceduras (such as ATP 16-26-00712 for the Indian Point 2 PO 9874769 D9421, covering bridge M1 cable P
$/N 00001 -- data taken 13,1988)/N 16-26-00712-5,k testing only af ter connector October cover helium lea attachment. This reflects the hermetic seal capability of the l
Whittaker connectors now used. Of the P0s reviewed by the inspec-tor, only the 1984 Turkey Point 4 order recognized the conflict between Specification 6102, Revision 4 and post 1984 actual i
practice. TechnicalChangeRequest(TCR)No. 9470282-2, dated 3
March 13, 1984, to PO 940282-16081 states that Specification 6102, Revision 4 paragraph 2.3.1 should be changed to read as follows: *The connectors used to terminate the ends of the cable specified in this specification and applicable drawin purchaser supplied if not provided by the sup liar." gs shall be However l
even the Turkey Point 4 order did not relax t e requirement of f
specification paragraph 4.2.5 for separate cable end seals.
i Failure to update Specification 6102, Revision 4 thus caused two I
types of conflicts in PCs involving Whittaker manufactured connectors.
First, paragraph 2.3.1 conflicted with the CE l
assembly drawing referenced in the PO regarding the source of i
connectors. Second, paragraphs 2.3.1 and 4.2.5 conflicted with what both CE and Whittaker intended for both connector types and cable leak test requirements. The affected P0s reviewed by the inspector are Turkey Point 4 (1984, leak test only), South Texas l
1 and 2 Salem 1 and 2 and Indian Point 2.
While no corrective action Is required for,the hardware shipped to these facilities, t
CE advised that these discrepancies are aeing addressed in a specification revision currently in progress, and had been noted in an internal QA audit.
1 The inspector questioned CE engineering regarding the use of shrink tubing and potting materials for either hermetic or t
environmental seals in CRT and RVLM systems.
Neither type of seal material is used for either purpose in CE-supplied systems built by Whittaker.
Both are used in various applications to restrict notion, provide strain relief, fill voids to prevent moisture collection, and the like.
The inspector questioned CE concerning their audits of Whittaker.
The CE Quality Operations Department has audited Whittaker in 1983, 1984, 1986, 1987, and twice in 1988. Whittaker was i
-....--e cw v.~,,-
---,,-w.--.-~-c...-.m.._,,,,,..--,~.e--,,,,-,me,-,,,,-,
--.--...-.-~-.--w..r.---..--
I i
)
ORGANIZATION: COMBU$ TION ENGINEERING, INC.
l WIND 50R, CONNECTICUT I
REPORT INSPECTION NO.: 99900401/89 01 RESULTS:
PAGE 5 of 6 removed from CE's Approved Suppliers List in 1988 and CE has subsequently performed 100 percent source inspection of all Whittaker recently appointed a new Quality Assurance
)
parts.
lQA) manager and CE is reviewing a revision of Whittaker's QA
{
manual.
I 1
The inspector noted another instance of apparent Whtttaker QA l
nonconformances in the CE files reviewed a CE le.tter referred to connectorsbuilt(andshipped)toanunapproveddrawing, j
CE's control of procurement activities will be carried as an t
Unresolved Itent to be addressed in a future NRC inspection.
In the interim an NRC inspection of Whittaker is planned which will p(99900401/89-01-01)rovide additional information on the subject.
l 2.
Part 21 Notification on Control Elements Assembly (CEA) Slippsce I
On April 20,lippage of two CEAs in different groups in Palo1 concerning s Verde 1.
The inspector interviewed CE engineering personnel and reviewed a CE Nuclear Safety Comittee review nemo dated April 21, 1989.
In conjunction with the licensee, Arizona Nuclear Power i
Project (ANPP), CE developed an explanation for the event.
The event is believed to be restricted to plants using the four coil CE drive motor (CEDM) design unique to the three Palo Verde units and WNP-3 (which is not yet operating).
During assembly the insulation of the lower lift coil lead wires is believed to be damaged by an inside thread as the wires are fed through a hole.
The threads are subsequently covered by a nipple, preven-ting further abrasion of the insulation. The lower lift coil is unique to the four coil CEDM, and is the only coil believed to move during rod motion.
Intermittent shorting of the damaged coil lead wires to the nipple is believed to be the source of electrical noise that adversely affects the rod control system.
i Evidence supporting this hypothesis includes a plant history of intermittent ground faults and evidence of damage including arcing in two lower lift coils removed from Palo Verde 2.
Each of the 88 rods in a CE core has a three diode controller fed by a 240 volt three phase bus comon to all 88 rods. Each phase has a zero cross detector that senses when the phase i
t i
i ORGANIZATION: COM80$T!0N ENGINEERINC, INC.
WINDSOR, CONNECTICUT REPORT INSPECTION NO.
99900401/89-01 RESULT 5:
PAGE 6 of 6 I
voltage crosses zero and triggers gate circuits for the diode l
controllers, Adjustable delays in the este circuits for each individual rod control the actual dc voltage applied to the CEDM coils. Since the noise caused by intersiittent shorting of a defective lift coil is introduced ir.to the comenon three phase i
t bus, the rod with the lowest CRDM voltage based on the gate circuit delay adjustments is most likely to be aff,ected by noise
{
anywhere in the power system. Thus in addition to possible i
slippage of the rod experiencing intermittent short circuits, i
one or more additional rods without regard for group assignments could also slip. Testing b supports this explanation. y CE with low resistance ground paths l
The analysis described above indicates that an intermittent fault in the lower lift coil can cause the affected rod to slip and can i
cause slippage of one or more additional rods. Although similar rod control systems are used in other plants, the four coil CEDM l
design appears to be a necessary contributor and that design is used only at Palo Verde and WNP-3.
Final resolution of the problem depends on further analysis and testing by ANPP and CE. The inspector concluded that analysis to date supported the 10 CFR Part 21 report in restricting the probable scope to Palo Verde and WhP-3.
j F.
PERSONS CONTACTED
+W. A. Goodwin, Director, Technical Services and Products
- A. E. Scherer, Director, Nuclear Licensing j
- +S. A. Toelle, Manager Nuclear Licensinn
}
- +0. M. Betancourt, Senior Consultant, L' censing j
- +J. M. Burger, Manager, Reactor Mechanical Systems M. J. Linden, Senior Engineer, Reactor Mechanical Systems l
+R. J. Fitzgerald Director, Quality Operations i
S. L. Mara, Senior Engineer, Quality Operations
- C. W. Ruoss Manager, Mechanical Engineerin
- W. R. HudnaII. Task Manager (CEDM Control) g and Technology l
' Attended entrance meeting on May 22, 1989
+ Attended exit meeting on May 25, 1989 b
h g
1 f