Having outlined the main concepts underlying the AGR fuel cycle, brief reference will now be made to some of the practicalities of trying to ensure that the principles are adhered to on the operating reactor. It is all very well to conceive the ideal initial loading pattern and subsequent optimum refuelling strategy, but for one reason or another it sometimes becomes necessary to depart from what could be termed ‘nor­mal fuel cycle practice’. For example, the design fuel cycle at Hinkley Point В and Hunterston В assumed that refuelling would take place continuously at a rate of one exchange every four or five days whilst the reactor remained at full power, This has not been possible in practice and a very significant proportion of the earlier re fuellings were carried out in large batches, perhaps 8, 12 or even 20 stringers at a time, with the reactor shutdown. Furthermore, when low power refuelling became possible the batch-refuelling regime was still retained. Such operational practice led to a re-examination of the earlier studies, using the sophisticated techniques of 2- and 3-dimensional com­puter modelling of core physics (i. e., enrichments, burn-up, reactivity, form factor, etc.), and revised schemes of refuelling were evolved based on the batch — refuelling idea. These took into account the different fuel irradiation and therefore reactivity distributions which were produced within the reactor both during the approach to equilibrium and in the later charges.

As we have seen, the Hinkley Point and Hunterston AGRs originally worked to an 18 GWd/t fuel cycle, but currently the reactors are moving away from an 18 GWd/t equilibrium condition to one at 21 GWd/t using fuel of new feed enrichments and with burnable poison. Such a transition also required a completely new study of the fuel cycle with yet another revised optimum refuelling strategy resulting.

The determination of the detailed sequence of re­fuelling is one of the most important end-products of a fuel cycle study and compliance with it is vital to optimum reactor performance both in the short and long term. It predetermines the sequence of re­fuelling to be followed in all the replacement charges at equilibrium and, therefore, whatever the fuel irra­diation discharge limit and irrespective of whether continuous or hatch refuelling is assumed, it is ex­tremely important that the sequence is rigidly followed in practice. Nevertheless there are a few occasions when this cannot be honoured, such as the need to prematurely discharge for a stringer containing ex­perimental fuel (post irradiation examination) in the generic interests of fuel design and development. It is also possible that fuel may fail in service, therefore requiring urgent removal in order to restrict the spread of contamination within the reactor. In both these instances the exchange of relatively ‘young’ fuel for new would not be conducive to either optimum fuel utilisation or reactor performance, and the discon­tinuity produced in the refuelling sequence would re­main in all later charges.