Reactor gas inlet temperature can be assumed, for the time being, to remain constant. If a change in gas flow occurs, for example, because of a blower/circulator failure, an increase in fuel temperature will occur. The negative fuel temperature coefficient of reactivity will cause reactivity to become negative so neutron power will decrease, this is an advantage because it helps to arrest the rise in fuel temperature. The increase in fuel temperature is transmitted to an increase in channel gas temperature and, where an auto control system is fitted and in use, the regulating rods will be driven in by the control system. This action will cause neutron power to be further de­creased until channel gas outlet temperatures are re­stored to the demanded value, and at this stage the reduction in reactor power will be just sufficient to match the reduction in gas flow.

Auto control systems on channel gas outlet tem­peratures are normally designed to cope with faults such as blower/circulator failure, so that there is little required of the reactor control engineer in the short term (as far as reactor parameters are concerned, although he has several operations to carry out on the failed blower/circulator and its associated boiler) other than to check the correct operation of the auto control system. This is the case on stations with six or eight blowers/circulators per reactor, where the loss of gas How amounts to one-sixth or one-eighth of the total, and on AGRs where the characteristics of the reactor and the more sophisticated control and protection systems allow rapid rates of change of reactivity. However, on magnox reactors with four blowers per reactor and on those with once-through boilers, prompt action by the reactor control engi­neer to assist the auto control system by insertion of manually-operated rods is probably required. Such action is also helpful on the six — and eight-blower sta­tions; on stations without auto control or where the auto control is not in use then prompt operator action is essential to prevent an undue rise in temperature.

So far we have considered only the first few min­utes of the transient. As time proceeds any remaining mismatch between the actual and desired power will cause the moderator temperature in a magnox reactor to change. Moderator temperature changes slowly, so an auto control system can readily cope with such changes. If the mismatch is such that reactor tem­peratures are too high, control rods can readily be inserted to restore the desired temperatures, if how­ever the moderator temperature is too low, for ex­ample, because of over-insertion of manually operated control rods, problems can arise either where the auto regulating rods are running only lightly inserted into the core or where no auto control is available. In these cases the rate of reactivity release from control rod withdrawal is small, and the reactor may tend to shut down on ‘temperature poisoning’. Further action which may be taken by the reactor control engineer to recover the situation is described in Section 5.5.4 of this chapter. Good operator training minimises the risks of such a problem occurring, particularly as the high reliability of plant in recent years has meant that such faults are rare and reactor control engineers may not experience such a fault on the plant for many years.

If the reduction in gas flow and reactor power are sustained for a period of hours, changes in Xe-135 concentration must be considered. As neutron power is reduced the xenon concentration increases (see Sec­tion 2 of this chapter), adding negative reactivity to the core. The rate of change of reactivity produced by this effect is slow and therefore compensation is easily achieved initially by rod withdrawal. However the total reactivity change, reached about six hours after the start of the transient, will be large if the neutron power change is large, and there may be insufficient rod worth available to offset the change in xenon worth and maintain criticality on the re­actor. Further action which may be taken by the reactor control engineer to provide additional posi­tive reactivity to offset the xenon change is described in Section 5.5.4 of this chapter. If sufficient positive reactivity cannot be found to offset the increase in xenon worth, the reactor will tend to shut down on ‘xenon poisoning’.

So far it has been assumed that a rise in reactor gas outlet temperature above the pre-fault value can­not be tolerated. However, if the pre-fault value is below the limits set by the Operating Rules, for ex­ample, in a magnox reactor if the reactor is operating to corrosion control limits which are more restric­tive than the Operating Rule limits, then a short-term increase in reactor gas outlet temperature up to the limits set by the Operating Rules is tolerable. In this case if it is found that the reactor is shutting down due to ‘poisoning out’, the reactor control engineer may consider it worthwhile to further reduce gas flow by adjustment on the healthy gas circuits, thereby inducing an increase in moderator temperature. This option is not available on an AGR where the mod­erator temperature is held constant by the re-entrant gas flow. It is a difficult operation because of the risk of tripping on excess temperature and because two variables are being changed at the same time, also because the control rods are only lightly inserted into the core so they will be of little value in minute — to-minute control. However it is worth trying as a last resort if the reactor can be kept safely at power.