ML051100285

Part 21 Initial Notification and Preliminary Report - Cutler-Hammer A200 NEMA Size 1 Starters

ML051100285
Person / Time
Site:	Ginna
Issue date:	04/18/2005
From:	Willis W Spectrum Technologies Utilities Services USA
To:	Office of Nuclear Reactor Regulation
References
+KBR1SISP20050502
Download: ML051100285 (7)

Text

4-18-05; 4:22PM;SPECTRUM TECH. USA ;518 382 0283 # I/ 7 FAX CONTROL SHEET DATE: April 18.2005 Number of pages including this sheet 7 TO: Name: NRC Operations Center Phone #: (301) 816-5100 Fax #: (301) 816-5151 FROM: Name: Bin Willis Tel: (518) 382-0056 Fax: (518) 382-0283 COMMENTS: Part 21 Initial Notification and Preliminary Report - Cutler-Hammer A200 NEMA Size 1 Starters Equipment Identification: Starter, Non-Reversing, Size 1, 600VAC, 3 Pole, W/1 25VDC coil, Westinghouse/Cutler-Hammer P/N A200MI CS, Type B Thermal Overloads, Ambient Compensated, Manual Reset Only In March 2002, Spectrum Technologies provided 20 Class 1E safety related Cutler-Hammer A200 NEMA Size 1 starters to Rochester Gas & Electric Company - Ginna Station. We had purchased the starters as commercial grade items from Cutler-Hammer, and dedicated them per EPRI NP5652, method 1, Special Tests and Inspections. This dedication successfully verified the following critical characteristics:

• Markings

• Dimensions and Configuration

• Electrical Functional Attributes, Including:

Insulation Resistance Current Carrying Capacity Minimum Pickup and Drop Out Voltage Time Current Characteristics of Overload Relay Ginna Station advised us that they had installed one of these starters in the circuit for their Boric Acid Storage Tank (BAST) Heaters. Approximately eight (8) weeks later they received MCB Alarm B-3 1 for low A BAST temperature. Investigation found an open coil on the starter and replaced the starter. The failed starter was sent to the Ginna Materials Laboratory for analysis. This analysis indicates that a manufacturing defect caused the failure.

SCArrI 04/18/2005 NON 17:24 [TI/RX NO 99361 Ijooi

4-i1-O5; 4:22PM;SPECTRUM TECH. USA ;518 382 0283 # 2/ 7 The Ginna Materials Laboratory examined the starter using a low power binocular microscope for signs of defects, overheating, or cracks. Resistance measurements between the two coils in the starter were taken and they were then subjected to X-ray radiography. Subsequent to this, the coils were disassembled and sectioned to reveal the defect. The failed coil wire was also examined using scanning electron microscopy for signs of overheating or melting.

It was determined that the failure is the result of a manufacturing defect in the epoxy surrounding the coil.

The open circuit was caused when the epoxy cracked at the defect and broke a wire running through the epoxy. This wire connected the two coils and this failure caused the open circuit. The defect in the epoxy was caused by the tape surrounding the coils. A piece of the tape protruded into the area between the coils where the epoxy is poured. This created a weak area which resulted in the failure during service.

Figure 1 (attached) shows the starter as received and the coil after removal from the starter. No other defects were seen in the starter assembly except for in the failed coil.

Figure 2 (attached) shows: (A) an optical photograph of the coil before disassembly and sectioning, (B) a radiograph which identifies the crack location, and (C) an enlargement of the radiograph showing where the wire failed.

Figure 3 (attached) shows: (A) an optical photograph of the coil with one plastic cover removed. (B) the coil cross section. The coil was constructed by inserting the copper wire (wrapped in yellow tape) into the housing. Epoxy was then poured into the housing and hardened in stages. (C) shows the epoxy surface with the crack. Some of the copper wire and yellow tape has been removed to show the crack.. Figure 3C shows the epoxy surface with the crack.

The epoxy also separates the two coils except for the copper wire running between them. The yellow areas seen in Figures 4A and 4C are where the yellow tape is sticking out into the epoxy. Figure 4A shows the cracked epoxy surface. The yellow are on the surface is where the tape protrudes into the epoxy. Figure 4B shows the surface of the fracture where the copper wire broke. Figure 4C is a scanning electron microscope photograph of the fracture surface showing the yellow tape. This is the area where the fracture originated. Figure 4D shows an optical photograph of the fracture origin. The fracture surface appears brittle and lines radiating away from the origin can be seen.

No signs of overheating, tackiness or discoloration are visible.

It is likely that the epoxy did not crack until the part was placed in service. Since the epoxy between the two coils was 90% penetrated by the tape, any normal vibration or heat cycling may have been enough to cause the crack.

Ginna plans to examine any additional pre-mature failures forpresence of cracks. They are also radiographing selected starters from the same lot for the presence of defects and will take additional actions based upon their findings or based upon the manufacturer's recommendations.

04/18/2005 RON 17:24 [TX/RX NO 9936] 1002

4-18-05; 4:22PM;SPECTRUM TECH. USA  ; 5 18 382 0283 # 3/ 7 We have provided the above information, including all figures, to Cutler-Hammer for them to investigate/evaluate the situation. We have requested that they provide us with a formal report addressing what the cause of the condition is, whether it is unique or a common problem, what other components may be affected, and what will be done to prevent recurrence. We are expediting them for their response.

If you have any questions, please don't hesitate to call, fax, or e-mail us. If the quality of the faxed photographs is unacceptable, please provide an e-mail address and color copies will be provided.

Assuring you of our best intentions.

William K Willis Vice President, QA Spectrum Technologies, A Division of ATC 112 Erie Blvd., Suite 3 Schenectady, NY 12305 Tel: (518) 382-0056 Fax: (518) 382-0283 cc: Mike Bauman, Ginna Station (585) 771-3884 04/18/2005 MON 17:24 [TX/RE NO 9936] I003

Figure 1 Contalor z

C C1)

D I

F Motor Control Coil I-2 a

U ELi Ca Co 2

04 i"

0l Cutler Hammer Motor Control U) a1 MolaWioi Laboatoy RE GinnW 3fI4 M

Figure 2 Coi Radiographs Z

Figure 2A -.i I -

Eirilargemeni 6 o!isr 213

.. (" "ro Aroa tbetvyatn ftq nva elIs

,v-

W

, -; ol lhe cal- Top View

I. _920, Figure 2C Area of Crack Coil Cal Copper Wire,%.

Area Beiween iht tkvo coils where Ihe wire RadograFh el fl Cal Top Viow Is broken Iioring flN wiresconneong exh cxii Crack-i-Figure 2Qis

.En enlargement t I ?,. ,

of 1his erea IF 1,

41 10radiogaphs mouifsy ol Ginna LIS grop A,*L -

.Crack RE ftoI~ MaW3i1a LaboratWr 3i1415.

90O0 [9C66 ON YU/flJ AZ:j. NOR SOO T/g1/iO Figure 3Coil Construction Yellow tapE wrapped " ox around coil xy fia Coll cross section Wt tape andcopper removd to showcrack RE trina Ma~eiials Laborairy 2JMM L. /9 # Sez0 Zee 815' VSn 'HOBOL fltJ1dLdS! WdZZ:t  !9O-8*-V

Craoked Epoxy Figure4 oil Frature Face I

iJ U

Lii a

(n An; .- __1 0

Fracture Surface with br6kdiYnwire Fracture surface wiere the incursion of tape caused a areainmte epoxy n

'K.CEGinn Maeud bo 3144