05000454/FIN-2009007-04
From kanterella
Revision as of 01:13, 16 November 2017 by StriderTol (talk | contribs) (Created page by program invented by StriderTol)
Finding | |
|---|---|
| Title | Insufficient Design Bases for Second-Level (Degraded) Voltage Timer Settings |
| Description | The inspectors identified an unresolved issue (URI) related to licensees failure to develop adequate design bases for the second level (degraded) voltage timer settings. Specifically, the licensee failed to evaluate the impact of operating and/or starting safety-related equipment at a voltage as low as 75 percent of the 4.16 kV nominal bus voltage for as long as 5 minutes and 40 seconds during an event involving a degraded grid voltage condition without a loss of coolant accident (LOCA) signal. The inspectors determined that the licensee did not have an analysis to demonstrate the ability of the safety-related loads to mitigate an event involving a degraded grid voltage condition when a LOCA signal was not present. Specifically, the inspectors found that, during a degraded grid voltage condition, if a LOCA signal was also present, after approximately ten seconds, the emergency diesel generators would start and accept the safety-related loads according to the prescribed load sequencing. However, if a LOCA signal was not present, the inspectors found that, after the ten-second delay, the degraded voltage condition resulted in an alarm in the control room and the start of a five-minute timer. The inspectors noted that Section A.4 of IEEE 741-1997, Degraded Voltage Relay Time Delay Settings, states, in part, that: After the voltage setpoint for the degraded voltage relays has been established, additional analysis is required to determine the appropriate time delays. These analyses will involve investigation of transient conditions, such as block motor starting and the effect of increased load currents from degraded voltage operation, on both protective device operation and equipment thermal damage. Two time delays should be determined by: a) the first time delay should be of a duration that establishes the existence of a sustained degraded voltage condition (i.e., longer than a motor starting transient). Following this delay, an alarm in the control room should alert the operator to the degraded condition; and b) the second time delay should be of a limited duration such that the permanently connected Class 1E loads will not be damaged or become unavailable due to protective device actuation... Protective devices (i.e., circuit breakers, control fuses, etc.) for connected Class 1E loads should be evaluated to ensure that spurious tripping will not occur during this time delay period. Consideration should also be given for restarting/reaccelerating the loads, should transfer to the alternate or standby power source be required. Similarly, NUREG 0800, Branch Technical Position (BTP) 8-6 states: In addition to the undervoltage scheme provided to detect LOOP [loss of offsite power] at the Class 1E buses, a second level of undervoltage protection with time delay should be provided to protect the Class 1E equipment. This second level of undervoltage protection should satisfy the following criteria: a) The selection of undervoltage and time delay setpoints should be determined from an analysis of the voltage requirements of the Class 1E loads at all onsite system distribution levels and b) Two separate time delays should be selected for the second level of undervoltage protection based on the following conditions: i The first time delay should be long enough to establish the existence of a sustained degraded voltage condition (i.e., something longer than a motor-starting transient). Following this delay, an alarm in the control room should alert the operator to the degraded condition... ii. The second time delay should be limited to prevent damage to the permanently connected Class 1E loads... The bases and justification for such an action must be provided in support of the actual delay chosen. Functionally, the Byron degraded voltage protection was consistent with the recommendations of IEEE-741 and BTP 8-6 in that the design included two levels of undervoltage protection and two separate time delays for the degraded voltage condition. However, the inspectors noted that, while the licensee had developed an adequate justification for the setting of the undervoltage relays and the first time delay, the licensee had not developed a technical justification for the second time delay. The need for a full evaluation of degraded voltage conditions was originally identified by the NRC in 1976 and 1979 as a result of events at Millstone and Arkansas Nuclear One. These events and subsequent similar events were discussed in various NRC generic communication vehicles, including NUREG-0900-5 and Information Notices (INs) 79-04, 89-83, and more recently, IN 2000-06. In IN 89-83 the NRC described specific concerns with degraded voltage conditions and stated that, in the Millstone event, a grid voltage drop combined with voltage drops produced by the step-down transformers reduced the control power voltage within individual motor control centers and individual 480 Volt controllers to a level that was insufficient to actuate the main line controller contactors. As a result, when the motors were signaled to start, the contactor control power fuses were blown making several motors inoperable. As indicated previously, at Byron, a degraded voltage condition without a LOCA resulted in the undervoltage relays sounding an alarm in the control room and initiating a five minute timer. Based on the alarm response procedure, if the alarm was the result of a degraded voltage, the operators were required to call the grid operator to determine whether the grid voltage could be increased and monitor the bus voltage. If the voltage dropped below 75 percent, the operators were required to initiate a transfer of the loads to the emergency diesel generators. In comparison, with a LOCA present, the degraded voltage relays were set to automatically transfer the safety-related loads to the emergency diesel generators when the bus voltage dropped below 92.5 percent of the nominal voltage (4160 Volts). The inspectors were concerned that, if the voltage at the 4 kV bus dropped to slightly above 75 percent of the nominal voltage, the operating motors would experience approximately a 28 percent increase in current, also considering the design voltage of the motors (4000 V). If operated within the design limits and properly protected, these motors would most likely experience no major damage. During the intervening five minutes, however, the increase in motor load current could result in spurious breaker trips and the automatic restart of the same or redundant motors with consequent further decrease in system voltages. At the lower voltage buses, the voltage drop would be greater than 25 percent due to losses in step-down transformers, cables, and other interposing devices. This voltage drop, complicated by potential motor starts, including the potential start of the motor-driven auxiliary feedwater pump, if a plant trip occurred, could result in adverse consequences that the licensee had failed to evaluate. Discussions with the licensee regarding this issue indicated that the design was accepted by the NRC during the original review and provided a copy of the safety evaluation report (SER) issued by the NRC in February 1982. In the SER, it is stated that: ...if the degraded voltage is not corrected within 5 minutes, the bus will automatically disconnect from the offsite power source and connect to its onsite diesel generator. This is in conformance with the staff position and is, therefore, acceptable. Subsequently, in April 1989 following a meeting with the NRC to address the adequacy of the undervoltage protection scheme utilized at the Dresden Station, Commonwealth Edison (CECo) wrote to the NRC and committed to implement administrative controls and associated operator training, which directs the operator to immediately take action to disconnect safety buses if the 4160 Volt power supply drops below 75 percent of the nominal bus voltage... The objective of this procedure is to minimize to less than one minute, if possible, the time that safety-related motors and other equipment could experience severe undervoltage (below 75 percent) in the extremely unlikely event that such conditions are sustained for more than several seconds. This commitment was made for the five plants owned by CECo at the time of the meeting, including Byron. As in the SER case, the meeting minutes addressed only one variable, i.e., the minimum voltage level but not the duration. Therefore, it is not immediately evident that the NRC intended to accept a 75 percent voltage for five minutes. Furthermore, the meeting pertained to the Dresden plant and the design limitations may be different. The licensee was unable to produce any documentation that was provided to the NRC in support of their design/operation of the electrical system. The FSAR and the Technical Specification (TS) were consistent with the SER. They both acknowledged the existence of a five-minute timer, but neither the FSAR nor the TS bases addressed the voltage level at which the plants are allowed to operate for the specified period. In response to the NRC concerns the licensee issued IR No. 892610. In the IR, the licensee indicated that they would develop a technical basis for the five minute delay. In the interim, they were revising the alarm procedure to direct the operator to separate the emergency buses from the system auxiliary transformer, upon confirmation that a degraded bus voltage condition (below 92.5 percent) existed. This issue is considered unresolved pending: (1) evaluation of the licensees technical basis for the time delay between the on-set of a degraded voltage condition and the transfer to the diesel generators, without a safety injection (SI) signal; and (2) discussion with NRR to determine the licensing and design basis (URI 05000454/455/2009007-04 (DRS)) |
| Site: | Byron  |
|---|---|
| Report | IR 05000454/2009007 Section 1R21 |
| Date counted | Mar 31, 2009 (2009Q1) |
| Type: | URI: |
| cornerstone | Mitigating Systems |
| Identified by: | NRC identified |
| Inspection Procedure: | IP 71111.21 |
| Inspectors (proximate) | T Bilik B Metrow D Jones T Go V Meghani R Skokowski J Bishop R Ng C Thompson B Bartlett J Robbins B Josea Dunlop Z Falevits C Brown C Baron S Lewis J Gilliam A Stone A Della Greca |
| INPO aspect | |
| ' | |
Finding - Byron - IR 05000454/2009007 | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
Finding List (Byron) @ 2009Q1
Self-Identified List (Byron)
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||