Information Notice No. 83-18: Failures of the Undervoltage, Trip Function of Reactor Trip System Breakers

                                                            SSINS No.: 6835 
                                                            IN 83-18  

                                UNITED STATES
                       NUCLEAR REGULATORY COMMISSION 
                    OFFICE OF INSPECTION AND PROCEDURES  
                           WASHINGTON, D.C. 20555 

                                April 1, 1983

Information Notice No. 83-18:   FAILURES OF THE UNDERVOLTAGE, TRIP 
                                   FUNCTION OF REACTOR TRIP SYSTEM BREAKERS 

Addressees: 

All nuclear power reactor facilities holding an operating license (OL) or 
construction permit (CP). 

Purpose: 

This notice describes recent failures of reactor trip system circuit 
breakers with undervoltage (UV) trip attachments and provides additional 
information related to the UV attachment. 

Description of Circumstances: 

Reactor Trip System (RTS) breaker failures at Salem Unit 1 on February 22 
and 25 and RTS breaker failures at San Onofre Units 2 and 3 reported to the 
NRC on March 11, 1983 led to issuance of IE Bulletins (IEB) 83-01 and 83-04 
respectively. Results of testing of RTS breakers required by IE Bulletins 
83-01 and 83-04 have been reported to the NRC. These results show that 
electrical breakers may not be achieving the performance reliability 
expected of them. The problem apparently lies with the undervoltage trip 
attachment. Failures of Westinghouse DB type, Westinghouse DS type and 
General Electric AK-2 type RTS breakers have been reported. To our 
knowledge, all currently licensed PWRs which use breakers for the RTS use 
one of these types. However, other safety-related systems may be configured 
with these breakers or breakers of other manufacturers and types which use 
an UV attachment to perform the required safety function. Licensees which 
use these breakers in other applications should be aware of the likelihood 
of failure to trip via the UV device. It would be prudent to check the 
operation and maintenance of these breakers. 

IEB 83-01 was directed to plants using Westinghouse DB type breakers with 
undervoltage trip devices. A summary of the information submitted by 
licensees in response to the bulletin follows: 

(1)  No fairures of DB type breakers were reported as a result of the 
     requested tests. Only one unit, which is in an extended maintenance 
     outage involving ihe reactor protectioh system, has not yet performed 
     the required tests. 

(2)  Seven of the twenty-eight plants using the subject breakers had not 
     been maintaining the breakers per the recommendations in Westinghouse 
     NSD Data Letter 74-2, a copy of which was attached to the bulletin. 

8303040012  
.

                                                           IN 83-18 
                                                           April 1, 1983 
                                                           Page 2 of 4 

(3)  Some plants reported that they were using the guidelines contained in 
     the NSD Data Letter with certain exceptions (e.g., the maintenance 
     interval had been extended from the recommended 9 or 12 month interval 
     to either 18 months or during refueling) 

(4)  Because the lubricants recommended in the NSD Data Letter are no longer
     commercially available, some plants reported that they were,using 
     lubricants other than those recommended in the NSD Data Letter. 

IEB 83-04 was directed to plants using RTS breakers with undervoltage trip 
devices, other than those using DB type breakers, such as General Electric 
AK-2 type breakers and the Westinghouse DS-416 type breakers. A summary of 
the information submitted to date by licensees in response to IE Bulletin 
83-04 follows: 

(1)  Several failures of the General Electric AK-2 type breakers and the 
     Westinghouse DS-416 type breakers were experienced during the tests and
     were reported to the NRC as required by the bulletin. Failures 
     involving the AK-2 type breakers were reported at the Maine Yankee 
     plant and at the Calvert Cliffs Unit 1 and 2 plants. On March 15, 1983, 
     Maine Yankee reported that three of the eight RTS breakers used at the 
     plant failed to open in the required times; one breaker tripped in 1.9 
     seconds, another in 2.7 seconds, and the third in 10.3 seconds. Maine 
     Yankee further reported that these breakers were subsequently cleaned 
     and lubricated after which they operated within the acceptable normal 
     time. On March 15, Calvert Cliffs reported that two of the eight 
     breakers in Unit 1 required approximately 6 seconds to trip and that 
     two of the eight breakers in Unit 2 required approximately 1.5 seconds 
     to trip. In its written report to the NRC dated March 21, 1983, Calvert 
     Cliffs described three additional failures that were detected during 
     tests at Unit 1 and three additional breaker failures detected at Unit 
     2. The report also indicated that these breakers functioned within the 
     acceptable response time subsequent to performing maintenance on the 
     breakers. On March 18, McGuire Unit 2 reported that one RTS breaker (DS 
     type) failed to open upon deenergizing the undervoltage coil and on 
     March 19, McGuire Unit 1 reported a failure of one RTS breaker. On 
     March 24, Arkansas Nuclear One (ANO) Unit 1 reported failure to trip of 
     an AK-2 type breaker.  

(2)  Failures not previously reported to the NRC and identified thus far in 
     the bulletin responses include the following: ANO Unit 1 identified one
     RTS breaker that failed to open and which was subsequently tested 10 
     times successfully after which it was surveillance tested daily until 
     the breaker was ultimately replaced. McGuire Unit 2 identified a 
     breaker failure that occurred during preoperational testing after which 
     it was reworked and placed into service on February 18, 1983 (this is 
     the same breaker for which we received a failure notification on March 
     18). Calvert Cliffs identified four events involving the undervoltage 
     trip device. Two of the events occurred on Unit 2 and were described as 
     sluggish trips which did achieve the protective system trip function. 
     The other two events occurred one on each unit and resulted in the 
     failure of the undervoltage device to trip the breaker. Three Mile 
     Island Unit 1 identified a failure which occurred on November 19, 1976,
     during post 
.

                                                           IN 83-18  
                                                           April 1, 1983  
                                                           Page 3 of 4 

     maintenance testing and which was stated to be due to binding caused by
     transporting and installing the breaker. Oconee, Unit 3 identified one 
     failure which occurred on December 17, 1978, for which the failure 
     mechanism was not determined and the breaker was replaced. 

It should be noted that IEB 83-01 failed to request information on previous 
failures of DB type RTS breakers which had not been reported to the NRC, as 
IEB 83-04 did for other type RTS breakers. Any information on previously 
unreported events of sluggish operation of all breaker types would be useful 
to the NRC. 

The information from these plants shows that in some instances breaker trip 
operation and timing may be erratic when breakers are tested by tripping the 
UV trip attachment. Although all of the GE AK-2 breakers did trip at Calvert 
Cliffs, the breaker trip times varied significantly from one test to the 
next. At McGuire, a Westinghouse DS-416 breaker intermittently failed to 
trip on successive tests. At ANO 1, an AK-2 breaker which tested 
successfully in response to the IE bulletin, failed a few days later during 
unrelated activities. The root cause of these failures has not yet been 
determined. 

NRC meetings and discussions with various regulatory review groups, 
licensees, and vendors have pointed out the importance of certain aspects of 
breaker operation, maintenance and design. The breaker tests at Calvert 
Cliffs 1 and 2 and Maine Yankee showed that although all of the AK-2 
breakers did trip, the trip times of some breakers varied significantly from 
one test to the next, up to a maximum of approximately 25 seconds. This type 
of sluggish operation is not acceptable. There are some indications that if 
the pick-up voltage for the AK-2 UV device is not set accurately, sluggish 
and erratic response times may occur. It is our understanding that the 
proper response time of the AK-2 breaker is less than 100 milliseconds. 
General Electric has described problems with lubricants used in breakers 
which become sticky after about 100 months of exposure. RTS breaker tests of 
the UV attachment which measure trip times may be a means of detecting a 
precursor of failure of the breaker to trip. 

Preliminary study of breaker problems indicates the necessity of regular, 
careful maintenance of RTS breakers. To reduce the likelihood of RTS breaker 
failure, Combustion Engineering and Babcox and Wilcox in cooperation with 
General Electric), and Westinghouse have developed updated maintenance 
procedures which are expected to be available within a short time. 

The NRC has met with both General Electric and Westinghouse and learned that 
there may be limitations associated with the design life of the UV devices 
and with the low design margin in terms of the torque available to trip the 
breaker by the UV device. Periodic replacement of UV devices may be 
necessary due to wear. Consultation with the vendor is recommended. The 
torque available to trip th,e General Electric AK-2 breaker by the UV device 
is critical to proper operation. General Electric has indicated that breaker 
trip bar torque measurements are needed periodically (quarterly) to detect 
the onset of problems. 

Preliminary review of the Salem event has pointed to one other area of plant 
operations that is important for detecting RTS breaker operational 
degradation and for preventing failures. The Salem precursor event of 
February 22, 1983,  
.

                                                           IN 83-18 
                                                           April 1, 1983 
                                                           Page 4 of 4 

was not recognized as a failure until after the February 25 event. A 
thorough post-trip analysis including close scrutiny of the events recorder, 
could have detected the problem associated with RTS breakers. 

It is suggested that holders of operating licenses or construction permits 
review this information for applicability to their facilities. 

No written response to this notice is required. If you have any questions 
regarding this matter, please contact the Regional Administrator of the 
appropriate Regional Office, or this office. 



                              Edward L. Jordan, Direcotr 
                              Division of Emergency Preparedness 
                                and Engneering Response 
                              Office of Inspection and Enforcement 

Technical Contact:  G. Lanik, IE 
                    (301) 492-9636 

Attachment: 
List of Recently Issued Information Notices
.