ML18065A995

LER 96-002-01:on 960116,initiated TS Required Shutdown Due to Safeguards Cable Fault.Both Sets (Six Cables) of Cables Were Replaced & Installed Through Turbine Generator Bldg

ML18065A995
Person / Time
Site:	Palisades
Issue date:	10/04/1996
From:	Mathews C CONSUMERS ENERGY CO. (FORMERLY CONSUMERS POWER CO.)
To:
Shared Package
ML18065A994	List: ML18065A993 ML18065A995
References
LER-96-002-01, LER-96-2-1, NUDOCS 9610090064
Download: ML18065A995 (6)

Text

• NRC FORM366 (4195) U.S. NUCLEAR REGULATORY COMMISSION APPROVED BY OMB NO. 3150-0104 EXPIRES 4130/98 LICENSEE EVENT REPORT (LER) (See reverse for required number of digits/characters for each block) ESTlllATED IUUlEN PER RESPONSE TO COllPL v WITH nas llNIDATORV INFORMATION COUECTION REQUEST: 50.0 HRS. REPORTED LESSONS LEARNED ARE INCORPORATED INTO THE LICENSING PROCESS AND FED BACK TO INDUSTRY.

FORWARD COllllENTS REGARDING BURDEN ESTIMATE TO THE INFORMATION AND RECORDS IWW>EllENT BAANCH (T.e F33), U.S. NJCLEAR REGUlATORV COlllllSSION, WASHINGTON, DC 20555-0001, AND TO THE PAPERWORK REDUCTION PROJECT (315().(1104, OFFICE OF llNWlEllENT AND BUDGET, WASHINGTON, DC 20503 FACILITY NAME ( CONSUMERS POWER COMPANY -PALISADES PLANT DOCKET NUMBER (2) 05000255 Page (3) 1of6 TITLE (4) Licensee Event Report 96-002-01, Initiation of Technical Specifications Required Shutdown Due to Safeguards Cable Fault -Supplemental Report EVENT DATE (5) MONTH DAY YEAR 01 16 96 LER NUMBER (6) YEAR I SEQUENTIAL NUMBER 96 002 REVISION NUMBER 01 REPORT DATE (7) MONTH DAY YEAR 10 04 96 OTHER FACILITIES INVOLVED (8) FACILITY NAME DOCKET NUMBER FACILITY NAME 05000 DOCKET NUMBER 05000 OPERATING THIS REPORT IS SUBMITTED PURSUANT TO THE REQUIREMENTS OF 10 CFR§: (Check one or more) (11) MODE (9) 20.2201(b) 20.2203(a)(2)(v) x 50.73(a)(2)(i) 50.73(a)(2)(iii)

I POWER I . LEVEL (10) 20.2203(a)(2)(i) 20.2203(a)(3)(ii) 50.73(a)(2)(iii) 73.71 . LICENSEE CONTACT FOR THIS LER (12) NAME Clayton M. Mathews, Licensing Engineer TELEPHONE NUMBER (Include Area Code) (616) 76.4-2305 COMPLETE ONE LINE FOR EACH COMPONENT FAILURE DESCRIBED IN THIS REPORT (13} CAUSE SYSTEM COMPONENT MANUFACTURER REPORTABLE.

CAUSE SYSTEM COMPONENT MANUFACTURER REPORTABLE TO NPRDS TO NPRDS B EA CBL X999 N SUPPLEMENTAL REPORT EXPECTED (14) I YES I NO If ves COMPLETE EXPECTED COMPLETION DATE x EXPECTED SUBMISSION DATE (151 (Limit toJ4Q0 f9 >e., approximately 15 single-spaced lines) (1-6) MONTH DAY YEAR On January 16, 1996, at 0536, with the plant operating at 100% power, the 2400V AC Safeguards Bus experienced a phase-to-phase fault. This fault initiated protective relaying which fast transferred buses 1 C, 1 D and 1 E from *safeguards

• powerto start:.:up

• power.* The plant resp-onseftcf

-the phase-to-phase fault and resultant fast transfer was as designed.

The fault caused the loss of the safeguards power source and placed the plant in a 24 hour2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> Technical Specifications action statement.

The reactor was shut down and the plant cooled down to less than 300°F within the time required by Technical Specifications.

/ The phase-to-phase fault was determined to be associated with the feeder going from the Safeguards Bus to the bus 1 D supply breaker. The cables were replaced, tested and declared operable.

Samples of the faulted cable were sent to laboratories for testing and analysis.

Analysis revealed that the cable failure was caused by localized water and contaminant treeing of the cable insulation.

961C090064 961004 PDR ADOCK 05000255 S PDR

• NRC FORM 366a 4195 U.S. NUCLEAR REGULATORY COMMISSION LICENSEE EVENT REPORT (LER) TEXT CONTINUATION FACILITY NAME <1 \ CONSUMERS POWER COMPANY PALISADES NUCLEAR PLANT DOCKETl2\

05000255 TEXT (If more space is required, use additional copies of NRC Form 366A) (17) EVENT DESCRIPTION LER NUMBER 16\ YEAR I SEQUENTIAL REVISION NUMBER NUMBER 96 002 01 PAGE £3\ 20F6 On January 16, 1996, at 0536, with the plant operating at 100% power, the 2400V AC Safeguards Bus experienced a phase-to-phase fault. The Safeguards Bus protective relaying activated the differential relay on the Y phase. Also activated were the ground relays and alarms for buses 1C,. 1 D, and 1 E. The result was a fast transfer of buses 1 C, 1 D, and 1 E to start-up power. The to-phase fault and subsequent fast transfer actuated the first level of undervoltage relays. The two faulted phases were the Z and X phases, as documented by the undervoltage relays and subsequent megger testing. A review of the Technical Specifications and the Standing Orders was conducted.

A determination was made that a 24 hour2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> Limiting Condition for Operation (LCO) action for Station Power Transformer 1-2 was the most limiting condition.

When it became apparent that the problem could not be repaired, the plant was cooled down to less thar;i 300°F where the Technical Specification

3. 7 .1 was no longer applicable.

Both sets (six cables) of cable were replaced from the Safeguards Bus to 1 D bus. The new cables were installed through the turbine generator building.

An independent review concluded that the. modification was acceptable and in accordance with applicable codes, standards, and Palisades licensing requirements.

ANALYSIS OF EVENT A plan was dsvap* fb locate the fault. A short to ground was noted when meggering.the*feeder cables from the Safeguards Bus to the buses 1 C, 1 D, and 1 E. These cables were disconnected at

• the Safeguards Bus and meggered individually back to their respective switchgear.

This testing determined that the cables A 1203/A 12-X02/1 were shorted to ground* and that the fat1ltswere

• inside the underground duct bank conduit. The three cables parallel three other cables (A 1203/A 12-X02/2) that pass through another underground conduit. These six cables are the feeder cables from the Safeguards Bus to bus 1 D. When the A 1203/A 12-X02/1 cables were removed from the underground conduit, bare copper about the size of a quarter was noted on the Z phase at 156 feet from where the cables entered the underground duct in 1 C switchgear room. Then, at 172 feet, a 9 inch section of cable (on X phase) was exposed all the way around the cable. No other damage to the cable jacket or cables was noted. The area of damage to the cable was in a straight run of conduit inside the duct bank. The cables had been pulled from 1 C switchgear room toward the Safeguards Bus, so damage from cable pulling tension is not suspected.

If the cable and jacket had been damaged during installation,

• NRC FORM 366a 4195 U.S. NUCLEAR REGULATORY COMMISSION LICENSEE EVENT REPORT (LER) TEXT CONTINUATION FACILITY NAME l1 l CONSUMERS POWER COMPANY PALISADES NUCLEAR PLANT DOCKETl2l 05000255 TEXT (If more space is required, use additional copies of NRC Form 366A) (17) LER NUMBER 6\ YEAR I SEQUENTIAL NUMBER 96 002 -REVISION NUMBER 01 PAGE l3\ 30F6 areas of damage on the cable jacket would indicate that the cable had been mishandled.

Also, if the cable had been badly damaged during installation, the 16kV DC hypot test which was used at the installation time would have indicated low insulation resistance before the faults developed.

After the cables were removed from the conduit, a baroscope inspection of the conduit was conducted.

During this inspection, a lot of carbon buildup was noted in the conduit near where the cable was dama9ed. Markings indicating the presence of water could be seen in the conduit near where the cable was damaged, but there was no water in the conduit at the time of inspection.

The conduit did not seem to be damaged anywhere in a way to indicate that the conduit might have caused potential damage to the cable. Some rags were pulled through the conduit and all the items that were removed were collected in a bag. The items in the bag consisted of about one large handful of sand/dirt, a 518 inch rusted nut, and flat pieces of dried cement that had been removed from the bottom of the conduit. The flat pieces of concrete were liquid that had seeped into the conduit while the concrete was being poured around the conduit arid later dried in the bottom of the conduit. These flat pieces of dried concrete, which were about the size of a half dollar, were very sharp on the edges. There was no material indicating that any animals had been inside the conduit. A list of probable causes was developed by plant personnel and two industry experts .. One expert was from Sargent and Lundy and the other was from Consumers Power Company's Equipment Services Department.

Both of these .individuals have been involved in investigations evaluating cable failures.

AB iMf'igation summary report was developed by each expert. At that time, they considered the most likely causes to be: 1) chemical attack, 2) water breakdown, 3) defective cable provided by the manufacturer, or 4) a combination of the three items listed above. Sections of the cable were cut out as the faulted cable was removed and these sections were sent I off site to be evaluated by different testing laboratories.

These evaluations revealed that localized I water and contaminant treeing (treeing is a condition where microscopic voids in cable insulation I look like tree branches when a wafer-thin cable insulation sample is viewed under a microscope)

I initiated in the Ethylene Propylene cable insulation and progressed to failure. Localized foreign I * *.* matter (contaminants) and voids were found to be present in cable insulation samples. These I _ -*-. acted as initiating points for the treeing degradation.

I I Investigation confirmed that the cable insulation type was appropriate for the installed I configuration.

It is concluded that the cable failure was rarJdom. I NRC FORM 366a 4195 U.S. NUCLEAR REGULATORY COMMISSION LICENSEE EVENT REPORT (LER) TEXT CONTINUATION FACILITY NAME <1l CONSUMERS POWER COMPANY PALISADES NUCLEAR PLANT DOCKETl21 05000255 TEXT (If more space is required, use additional copies of NRC Form 366A) (17) Cable History LER NUMBER 16\ YEAR I SEQUENTIAL NUMBER 96 002 REVISION NUMBER 01 PAGE 131 40F6 The failed cables were installed in 1989 as part of a project to increase cable ampacity.

The cables were tested at the factory as part of a normal code requirement.

The tests were in accordance with ICEA S-68.,.516 and the IEEE Flame Test. The testing also included a factory hypot test of the cable. Following installation, the cables were hypot tested again with 16kV DC and found to be acceptable on 10/27 /89. Yellow 77 was used as a cable pulling lubricant in accordance with Palisades' cable installation specification.

The cable manufacturer advises against using this pulling compound because of the_ potential adverse reaction between the pulling compound and the cable semiconducting outer Yellow 77 is an oil based lubricant that penetrates the semiconducting jacket and removes . ,-, the carbon. Without the carbon to conduct, the jacket can become stressed which leads to premature failure of the cable. Howeve*r, initial iri'vestigation of the cable insulation not show any degradation that could be attributed to this lubricant.

Starting on 5/19/95, the plant began to experience bus ground alarms for buses 1 C, tD, and 1 E. Thirty momentary ground faults were documented on buses 1 C, 1 D, and 1 E from 5/19/95 through 9127195. No other grounds were experienced until the phase-to-phase fault on 1 /16/96 .. Meanwhile, the following actions were ongoing to locate this momentary ground: 1. A tempera:;

gm : :e was placed and removed when no effect was noticed; 2. An engineering consultant was asked to review and recommend an action plan. 3 Documentation was kept for equipment that was energized during a ground. 4. Cables and equipment that could be deenergized were meggered.

5. Plans were formulated to megger other cables (including the safeguards cables) and equipment during the 1996 refueling

_outage, which is scheduled to commence November 2, 1996. CAUSE OF THE EVENT The root cause of this event was that treeing initiated in the Ethylene Propylene cable insulation I and progressed to failure.

NRC FORM 366a U.S. NUCLEAR REGULATORY COMMISSION " 4195 ' LICENSEE EVENT REPORT (LER) TEXT CONTINUATION FACILITY NAME l1 l CONSUMERS POWER COMPANY PALISADES NUCLEAR PLANT DOCKETl2\

05000255 TEXT (If more space is required, use additional copies of NRC Form 366A) (17) SAFETY CONSEQUENCES AND IMPLICATIONS LER NUMBER 6\ YEAR I SEQUENTIAL NUMBER 96 002 REVISION NUMBER 01 . PAGEl3\ 50F6 Either the Safeguards Transformer or Station Power Transformer 1-2 supplies power to the Safeguards Bus. The Safeguards Bus, in turn, feeds power to Class 1 E buses 1 C and 1 D and non-vital bus 1 E. The Safeguards Bus supply to 1 C, 1 D and 1 E is redundant to the 2400V AC Startup Transformer 1-2 feed to buses 1 C, 1 D, and 1 E. The diesel generators are also available to supply buses 1 C and 1 D. Either startup power or diesel generators can supply normal and emergency power to necessary loads on buses 1 C and 1 D. With only one off site power supply available to feed the bus after the cable fault, and with the primary coolant system above 300 degrees Fahrenheit, the 24 hour2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> Limiting Condition for Operation (LCO) for the unavailability of power from the station power 1-2 became the most limiting condition.

Because the repairs could not be implemented before the LCO was exceeded, the primary coolant system was cooled down below 300 degrees Fahrenheit where the redundant power source was no longer requ_ired.

CORRECTIVE ACTIONS 1. Corrective Actions Taken and Results Achieved Both sets (six cables) of cable were replaced from the Safeguards Bus to 1 D bus. The new *cables were installed through the turbine building.

An independent review *-concluded that the modification was acceptable and in accordance with applicable codes, standards; Ji A &llsades licensing requirements.

• Megger and Electrical Circuit Characterization and Diagnostics (ECCAD) tests were -conducted on the other cables from the Safeguards Buswhich feed buses-1C and-1E. The *-ECCAD results indicated all cables were acceptable.

The megger results also indicated that all the cables were acceptable.

The cable pulling force and ampacity calculations for the faulted cables were recalcu.lated and both were found to be acceptable.

A review of the plant responses provided by logs and data logger' printouts was completed.

The conclusion of the review was that the plant equipment responded as designed.

• • ., '

• NRC FORM 366a U.S. NUCLEAR REGULATORY COMMISSION 4195 ' LICENSEE EVENT REPORT (LER) TEXT CONTINUATION FACILITY NAME l1 l CONSUMERS POWER COMPANY PALISADES NUCLEAR PLANT DOCKET/2\

05000255-' TEXT (If more space is required, use additional copies of NRC Form 366A) (17) LER NUMBER 6\ YEAR I SEQUENTIAL NUMBER 96 002 REVISION NUMBER 01 PAGE l3l 60F6 Preliminary results were presented to plant management and the Plant Review Committee (PRC) to assure all proposed actions that were needed prior to heatup would be identified and completed.

The PRC concluded that it was acceptable to return the plant to operation while the remaining testing and root cause evaluations were being performed.

The construction cable installation specification was revised to require the use of pulling lubricant specified by cable manufacturers.

Testing of the faulted cable from bus 1 D was performed to determine the cable failure mechanism.

Testing revealed that the cable failure was caused by localized water and contaminant treeing of the cable insulation.

The use of Yellow 77 was reviewed with the cable manufacturer (BICC). The concern with Yellow 77 involves a degradation of the semiconducting properties of the cable jacket. BICC supplied 1982 test data which shows an accelerated increase in jacket electrical resistance when in contact with Yellow 77. No evidence has been found that the properties of the jacket of the cable samples had been compromised.

.. The faulted cable test results have been evaluated and a determination has been made that no further action is required for existing plant cables that could experience common mode * . failure. The risk of having an in-service failure of other existing cables to be very low, and a1

• Z *ice fai_lure will be preceded by a ground alarm on the 2400 volt system. 2.

Actions to Avoid Recurrence None remaining. ( *1 I I I I I