Protecting People and the EnvironmentUNITED STATES NUCLEAR REGULATORY COMMISSION
ACCESSION #: 9712190171
SIEMENS
December 1, 1997
ABB Combustion Engineering Nuclear Operations
PO Box 500
2000 Day Hill Road
Windsor, CT 06095-0500
Attn: Jack R. Buckholz
RE: Failed MDR Relays at Entergy Operations, Inc.
Jack,
ABB has forwarded three reports from Southwest Research Institute to
Siemens Electromechanical Components pertaining to three failed MDR-7032
"medium" relays at Entergy Operations, Inc.'s Waterford 3 Nuclear
Generating Station. Included in these reports are an analysis and a
conclusion. Siemens Electromechanical Components agrees with the
analysis contained in these reports, however, Siemens Electromechanical
Components has reached a different conclusion based on its own analysis
and testing.
Siemens Electromechanical Components performed a number of temperature
tests to determine if the Nye Nyogel 718B grease degrades for the
possible temperature operations of these types of devices. Temperature
testing of bearings shows both bearings are approximately the same
temperature, A temperature test was conducted on the MDR-7068, the
highest wattage of the suspected relays. With a 10% over voltage of 33.0
VDC, the bearing near the mounting base checked 58 Degrees C, and the
bearing at the contact end checked 59 Degrees C Nye Nyogel 718B grease,
is rated by the manufacturer for 260 Degrees C, The bearings operate 200
Degrees C below the manufacturer max rating. A weight loss test of the
Nyogel grease is an on going test, After 696 hours at 150 Degrees C,
there was only 1.312% weight loss.
Siemens Electromechanical Components theory is the grease was
contaminated by silicon dioxide fibers. To test this theory, relays were
built using contaminated grease (test samples) and uncontaminated grease
(control samples) in the bearings. Relays were cycled for 162,000
operations then left energized continuously for 48 hours at 65 Degrees C,
The relays with uncontaminated grease still had fluid grease while the
units with contaminated grease was a thick paste. The conclusions of
these test shows that Nyogel grease does not degrade and cause the relay
to bind. The results clearly show that the contaminated grease Is the
root cause of the failure
Page 1
Siemens Electromechanical Components, Inc.
200 South Richland Creek Drive Tel 812 386-1000
Princeton, IN 47671-0001 Fax 812 386-2430
SIEMENS
Siemens Electromechanical Components did not analyze the failed relays at
Waterford 3 because these units were not submitted to us. The amount of
contamination that was within the relays is not known. Prior documented
history Return Material Evaluation number 34017 (attachment one) had 4
MDR-170-1 relays recalled from ABB due to contamination within the
relays, with date code of 9412. These 4 relays were the total order
built for ABB. The contamination contained silicon dioxide fibers from
contact ring insulator plates, flakes from the switch rings, and flakes
of the coil epoxy. When silicon dioxide fiber is mixed in grease it
becomes very abrasive, similar to lapping compound. This caused abnormal
wear particles from the shaft and bearings. The grease then absorbs the
wear particles and becomes too stiff for the relay to operate.
Only the medium relays listed in attachment 2 are suspected to have this
problem. There has been no reported failures on the small relays. The
date codes of the suspected medium MDR relays are from 9317 to 9532. The
reason 9317 is the suspected date codes is the start up date of the line
after moving from Princeton, IN to Marion, KY. Date code 9532 is the
initiation of a continuous improvement team. The team formalized
housekeeping procedures and added an enclosed booth that isolated MDR
final assembly from general production area preventing contamination.
The relay listed in attachment 2 with the date codes 9317 to 9532 can be
returned to the factory for inspection.
Respectfully submitted,
Dick Market
Staff Technical Specialist
Product Engineering
Attachments: RME #34017 and #34406, and Part list of Potential Affected
Devices.
XC: Dan Smith, SEC
Al Tessier, ABB
Page 2
Siemens Electromechanical Components, Inc.
200 South Richland Creek Drive Tel 812 386-1000
Princeton, IN 47671-0001 Fax 812 386-2430
SIEMENS
PROBLEM SOLVING AND RESOLUTION
Problem No. Title: Date Opened: 9/18/96 Assigned To:
34017 Contacts frozen First Rev. MDR Line
(1) Team Leader: Interim Problem Reported By: RME #:
Mark Bryant Final Customer 34017 & 34406
Date 9/18/96 Responsibility M. Bryant
(2) Describe Problem:
Customer: ABB Combustion Engineering Nuclear Operations
P.O. Box 500 M/S 9426-1936
Windsor, CT 06095-0500
Attn: Jack Buckholz
Product: MDR-170-1
Date(s): 9412 (red) (S/N 514)
Mfg. Plant: Marion
Customer References: CE Order# 407112-01, CE Reference# 011-95823-
0040
Siemens References: RA# 9643077, Ack# 13745, PO# 407112
RME 32276 (S/N 513), RME 34406 (S/N 511 & 512)
Reason For Customer
Return: Contacts frozen
Problem Identification/
Verification: Verified by Princeton Failure Analysis Lab,
rotor remains bound in energized position without coil power.
(2) Evaluation Date: 9/23/96 Responsibility: M. Bryant
Black substance rolled up around the tip of the rotor shaft on the bottom
bell. Substance is causing rotor to bind which prevents proper contact
operation.
(3) Containment & Interim
Corrective Action(s): Date: 9/18/96 Responsibility: M. Bryant
Contacted Sales to have Return Authorization issued for the remaining two
MDR-170-1 units built for the original release in 1994. S/N 511 & 512.
(RA 9643079 / RME 34406))
Due to the isolated nature MDR units presently in service or on the
shelf.
Siemens Electromechanical Components, Inc.
Potter & Brumfield 200 South Richland Creek Drive Tel 812
386-1000
Products Division Princeton, IN 47671-0001 Fax 812
386-2430
13E163 06/04/96
Page 2
RME # 34017
(4.) Define & Verify Root
Cause(Codes): Date: 12/18/96 Responsibility: M. Bryant
Materials & Processes Lab has determined the cause for bound rotor was
contamination of the bearing grease with glass particles and other debris
(S/N 514). The contaminants reacted with the grease to form a black
substance which would not allow the rotor to turn freely. The major
components of the contamination were silicon from glass fibers found in
the contact ring insulator plates and flakes of coil epoxy. These
components are inherent to parts handling in the assembly area M&P Lab
found similar contamination in S/N 511 & 512 although these units were
still operable. S/N 513 returned in October '95, should also be
considered to fall under this response (ref RA 9541066 / RME 32276)
Exactly how such a quantity of contaminants could accumulate in these 4
relays is not known. Due to line improvements, layout changes and
personnel turnover since 1994 it is difficult to identify a specific root
cause for this isolated problem.
M&P lab identification of the contaminants shows no indication this
problem is related to the coil varnish outgassing problem which caused
similar binding problems in the past (units built prior to date code
9024)
Coils were gaged on the MDR line and found to be well within tolerance
and in no way oversize.
Document changes required
Yes No N/A Date Changed
Change Flow Diagram X
Change Process FMEA X
Change Design FMEA X
Change Control Plan X
Change Attribute Procedure X
Routings X
Drawings X
Others X
13E163 06/04/96
RME # 34017 Page #3
(5.) Establish Corrective
Action Plan Date: 1/8/97 Responsibility: M. Bryant
While the source of the contamination to the motor assembly could not be
definitely determined, the following improvements have occurred since the
discrepant units were manufactured in 1994
- Enclosed booth built to isolate MDR final assembly stations from
general production area to reduce contamination and noise levels
- Added additional process step to fabrication of insulator plates Parts
are tumbled following blanking operation in order to remove loose
particles and fibers
- Initiation of Continuous Improvement teams and successful QS9000 effort
have resulted in formalized housekeeping procedures which mandate daily
clean up by the operators. These have contributed greatly to improving
general production area cleanliness
- Continuous improvement teams have also worked to significantly reduce
the inventory level of parts and sub assemblies stacked on the production
line and in the Stockroom The reduced inventory levels result in more of
a JIT system which limits the opportunity for contamination to the
subassemblies
A review of the present MDR production for further potential sources of
contamination showed the following problem areas:
- The assembly operators maintained a small cup of grease at each
assembly station and used a toothpick to apply a dab of grease from the
cup to the bearing surfaces. The grease cups were uncovered and would
collect contamination on the surface of the grease supply.
These cups were replaced with plastic syringes which seal off the grease
supply from any contaminants (eff 9640, 10/1/96)
- No instructions existed to prevent the operators from beginning
assembly of a unit and then later setting it aside to work on a different
unit, The possibility existed for a partially assembled unit to collect
contaminants on a greased bearing surface.
Operators have been instructed to ensure any partially assembled units
are properly covered when it becomes necessary to put them aside or they
are to complete the unit to the point of sealing the motor assembly
Grease is to be removed from any questionable units and fresh grease
applied. Instructions reflecting these requirements are in process to be
added to MDR Engineering Specifications (eff. 9640)
Existing assembly instructions require each motor assembly be blown out
with de-ionized air in order to remove any contamination prior to capping
the motor assembly with the too end bell. It is felt that the present
MDR assembly methods, process controls and housekeeping measures are
adequate to preclude further occurrence of this problem.
(6.) Implement & Verify
Corrective Action(s): Date: 1/8/97 Responsibility: M. Bryant
- 5 piece MDR sample taken from present MDR production and examined for
contamination, No contaminated motor assemblies were found.
- 6 MDR units successfully completed mechanical life testing to 500,000
operations @ 5 ops./min. on 12/22/95 This is submitted as an indication
of the reliability of recent production.
Date Implemented Date Verified: Verified By:
10/1/96 1/8/97 M. Bryant
Document changes required
Yes No N/A Date Changed
Change Flow Diagram X
Change Process FMEA X
Change Design FMEA X
Change Control Plan X
Change Attribute Procedure X
Routings X
Drawings X
Others X
13E163 06/04/96
RME # 34017 Page #4
(7) Actions to Prevent Recurrence: Date: 1/8/97 Responsibility M. Bryant
Actions taken are applicable to all MDR part numbers
(8.) Congratulate/Team Members: Date: 1/8/97 Responsibility M. Bryant
Distribution
Mike Carnahan
Gale Maxwell
Michael Goldhacker
Don Deel
Dick Market
Ed Wagner
Karen Greenwall
Jack Buckholtz ABB (FAX)
Document changes required
Yes No N/A Date Changed
Change Flow Diagram X
Change Process FMEA X
Change Design FMEA X
Change Control Plan X
Change Attribute Procedure X
Routings X
Drawings X
Others X
13E163 06/04/96
SIEMENS
Part List of Potential Affected Devices
With Date Codes of 9317 to 9532
MDR-141-1 MDR-6054 MDR-7033-SCE-0
MDR-141-1-SCE-2 MDR-6059 MDR-7034
MDR-141-2 MDR-6050 MDR-7034-SCE-0
MDR-142-1 MDR-6054 MDR-7033-SCE-1
MDR-167-1 MDR-6069 MDR-7036
MDR-170-1 MDR-45074 MDR-7045
MDR-170-1-SCE-2 MDR-6100 MDR-7048
MDR-170-2 MDR-6101 MDR-7052
MDR-172-1 MDR-66-4 MDR-7056
MDR-173-1 MDR-7020 MDR-7059
MDR-6007 MDR-7025 MDR-7067
MDR-6019 MDR-7029 MDR-7068
MDR-6041 MDR-7032 MDR-7069
MDR-60150 MDR-7032-SCE-0
MDR-7033
Attachment 2
Siemens Electromechanical Components, Inc.
200 South Richland, Creek Drive Tel 812 386 1000
Princeton, IN 47671-0001 Fax 812 386 2430
SIEMENS Siemens Electromechanical Components, Inc.
Fax
To: From:
Company: Department:
Department: Fax: (502) 965-1212
Fax: Telephone:
Telephone: Date: Pgs. to follow:
Message
Potter & Brumfield Products Division
726 Chapel Hill Road
Marion, KY 42064
502.965.3193
*** END OF DOCUMENT ***
