Fuel Re-ordering in the Bruce A Reactor Cores

Abstract

In the spring of 1993, physicists discovered that the power pulse due to a large inlet header break in the primary coolant system of the reactors could be higher than previously calculated and could potentially invalidate the safety analysis and licensing basis of the reactors. The higher power pulse was due to the fuel string relocation towards the channel coolant inlet. The amount of relocation had been increasing with operating time due to irradiation growth of the pressure tube. The safety assessments of the physics implications of the movement had not been considered in the analysis. The reactors were then de-rated to 60% of full power until the power pulse issue was resolved. Bruce A considered two design alternatives: reduce the fuel string axial gap, and reverse the fuelling direction. A reduction in the axial gap would reduce the size of the power increase. This solution was partially implemented by inserting the flow straightening inlet shield plugs (FSISPs) into the centre channels of the core. This had the immediate effect of decreasing the amount of relocation within the core during the postulated accident because the flow straightening shield plug is longer than the shield plug it replaced. This allowed a slight increase in operating power. A further gap reduction method was considered: the introduction of the long fuel bundle design, which was implemented at Bruce B. At Bruce A, the amount of benefit, and the foreseen difficulties in implementation led to the choice of reversing the fuelling direction, fuelling with the flow (FWF). This paper presents an overview of the steps in reversing the core and the problems encountered. The immediate concern with FWF was how to change the fuelling direction quickly and economically, and without causing the fuel to defect. A fuel shift scheme was devised in which 12 fuel bundles from one channel would be discharged into the fuelling machine in the normal direction, and then re-inserted into a neighbouring channel with the opposite flow direction. This would then place the high burnup fuel on the latches and the low burnup fuel at the coolant inlet. The fuel rearrangement would take place at power and only the lower burnup fuel would experience significant power increases. The next concern was whether the fuel could tolerate the reversal of the flow direction and the load applied to the irradiated bundle due to the latch. Test bundle irradiations showed that the fuel bundles would fail by delayed hydride cracking if also subjected to a thermal cooldown transient. Destructive inspections of the fuel in hot cells revealed cracks. New in-bay inspection tooling was devised which could examine the end plates using ultrasound technology and detect incipient cracks. The loads on the end plates were sufficiently high (this was confirmed by analysis) that a new design was necessary to accommodate fuelling with flow. A new fuel supporting outlet shield plug, F3SP, was designed to fully support the end plate through the latch and prevent cracking of the end plates. The shield plug was qualified in out-reactor loop tests. Test strings were reordered and the performance of the fuel was monitored by in-bay inspection. The core reversal of Bruce A units was in progress when the units were laid up in 1998.