Resolution of Generic Safety Issues: Issue 6: Separation of Control Rod from its Drive and BWR High Rod Worth Events (Rev. 1) ( NUREG-0933, Main Report with Supplements 1–34 )
BWRs are operated in such a manner as to attempt to produce an axial power distribution with as little peaking as possible. When the reactor moderator is not boiling (e.g., during startup and heatup), the axial flux distribution is much more peaked. This in turn causes incremental control rod worths to peak at a certain axial location due to the non-linear neutron importance weighting.
When a reactor is restarted shortly after a shutdown, the xenon distribution tends to enhance the peaking effect. This effect first came to attention when several BWRs experienced short periods when recovering from scrams. In some cases, reactors apparently went from subcritical to supercritical with 5-second periods by moving one control rod one 6-inch notch. In every case, the reactors scrammed long before power became significant. Nevertheless, short periods were considered detrimental to an operator's ability to safely control the reactor, and prevention of such incidents was addressed in IE Circular No. 77-075 which was later supplemented by IE Bulletin No. 79-12.6
The question then arose concerning implications for the analysis of the Rod Drop Accident (RDA). The ACRS described4 the concern as follows:
Separation of a BWR control rod, coupled with the rigidly controlled sequence in which such rods must be withdrawn, could have very serious safety implications. Thirteen rod separation events and six rod-worth-multiplication events (resulting in periods shorter than 5 seconds) have been reported since 1976. Previous studies have indicated that the probability of fuel damage from an RDA is small. This study suggests that the withdrawal of a prescribed rod whose worth has been magnified by high core xenon concentrations may be more probable than the erroneous withdrawal of a high worth rod under normal startup conditions. A determination needs to be made as to whether the high rod notch worth produced by high xenon concentrations extends over a large enough region of rod movement to cause fuel damage from a dropped rod. This study indicates that additional attention needs to be given to the manner in which control rods are withdrawn, particularly during conditions where rod worths deviate from the normal.
A study7 along the lines of that recommended by the ACRS was made by CPB/NRR. LERs were extensively studied together with the basic physics involved and the staff came to the following conclusions:
|(1)||The information in LERs on the separation of control rod and drive mechanisms does not indicate an increase in the probability of separation or of having an RDA event compared to conclusions from previous studies of this subject.|
|(2)||There is no reason to expect a significant control rod reactivity worth increase or RDA effect from xenon and no evidence from LERs on Intermediate Range Monitor (IRM) scrams that indicate that the expectation is incorrect.|
|(3)||Group-banked rod withdrawal has been and is used with several improvements in details of the Reduced Notch Worth Procedure (RNWP) to mitigate high notch worth effects.|
|(4)||There is no close approach to RDA limits in the xenon restart regime with GE SAR analyses and the margins are substantially increased (beyond any reasonable uncertainties in the rod worth analyses) when using appropriate thermal-hydraulic feedback in the analysis.|
Based on the results of the CPB/NRR study, it was concluded that no further action on this issue was required.382 Thus, this issue was RESOLVED with no new requirements.