For the detection and location of boiler leaks, a com­bination of a detector of the type described in the previous section and special sampling and indication techniques are used. The gas outlet of each individual boiler, pair of boilers or boiler quadrant, depending on the design, are sampled. A ‘first-up’ alarm system can be used to indicate which sample can be associated with the leaking boiler.

A differential system is also in use. Each signal is backed-off against the background or reference moisture concentration, for example, boiler inlet, to provide a differential output. This technique provides increased sensitivity by reducing zero drift. The sam­pling system is automated, the detector outputs being compared and processed by a microcomputer.

A recently developed system shown in Fig 2.69 has a pair of analysers for each of the four reactor quadrants, the objective being to determine which quadrant is associated with water ingress from a boiler leak.

The gas samples from a quadrant to its pair of analysers are periodically reversed, the outputs of the analysers are measured and the differentials cal­culated.

The action of the sampling valves, the measure­ments and calculations are under the control of a microprocessor, and it is anticipated that it will be
possible to determine a change of 5 vpm in a back­ground of 300 vpm.

The fuelling equipment to be described is that at Hinkley Point В which is very similar to that at Hunterston В, Heysham 2 and Torness. As Dungeness В and Hartlepool/Heysham 1 have differing equip­ment the major differences will be explained.

The original design fuel burn-up was IS 000 MWd/t ^ith the fuelling equipment designed for a lifetime corresponding to 1 OO’To load factor, 100% power at 12 000 MWd/t burn-up. This design fuelling rate corresponds to 4.5 fuel assemblies and one control rod assembly per week. Heysham 2/Torness have been redesigned for 24 000 MWd/t and a higher burn-up is being progressed for other stations.

7.1 New fuel facilities

The new fuel elements arrive by road transport stand­ing vertically in a nine-compartment base. The boxes are conveyed to the fuel store (see Fig 2.106). When required, a box is transferred to box opening. The elements are removed from the box by a grab which engages in the groove in the element outer sleeve. They are inspected and pressure tested since the graph­ite sleeve is subjected to an internal pressure differ­ence of some 2 atmospheres during charging at full load, and an over-pressure test will detect sleeves with unacceptable cracks.

Eight fuel elements together with the fuel stringer top and bottom assemblies are transferred to the new fuel assembly cell. The cell has a vertical assembly tube with a descending platform which is progressive­ly lowered as the fuel stringer is erected. The fuelling machine (FM) is positioned over the cell and contains a serviced plug unit complete with a new fuel tie bar. The cell operator, using a local hoist control, lowers the plug unit and guides the end of the tie bar into the fuel stringer. When the plug unit has been fully lowered to the cell lower level, the operator makes — off the tie bar with a special nut and collet. The new fuel assembly is now complete and is hoisted up into the fuelling machine. During the time the operator is in the upper part of the cell, a split shield in the roof is used to protect him from the highly radio­active lower plug unit; the tie bar, which is hanging from the plug unit, passing through a small hole in the shield. The split shield is only opened to allow the plug unit to pass through; this occurs after the operator has gone to the lower level where he is below additional fixed shielding.

At Dungeness В, the fuel stringer is assembled horizontally complete with its tie bar, spring tine coupling and other fitments in a split tube in a cra­dle. This is subsequently tilted to the vertical and the stringer is raised to engage the tine coupling Mth a serviced plug unit supported in the new fuel magazine.

At Hartlepool/Heysham 1, the fuel stringer assem­bly is also horizontal but with the top portion of the fuel tie bar projecting. The assembled stringer is raised to the vertical. The projecting tie bar is threaded through the neutron scatter plug of a ser­viced plug unit and coupled to the upper tie bar at the service joint across the neutron scatter plus.

The site layout is shown on Fig 2.124 with the per­manent works occupying some 16 hectares to the north of the existing A station. The site layout has been designed on the basis of the following require­ments and economic considerations:

• Functional relationship of the main power block and ancillary buildings.

• Architecture, landscaping and maintaining the local environment.

• The connections to existing overhead power lines on the western side of the site.

• Considerations of lengths of CW culverts and main cable routes.

• Operational access and personnel flow.

• Separation of plant for reasons of safety.

Referring to Fig 2.125, the main power block, which houses the main power generation plant, comprises the following buildings located in a central group:

Reactor building {and secondary containment)

Auxiliary building

Control building

Fuel building

Mechanical annexe

Turbine house

Other principal functional buildings are:

Essential diesel building

Auxiliary shutdown and diesel building

Radwaste process and storage building

Auxiliary boiler house

CW pumphouse

Hydrogen generation complex

Reserve ultimate heat sink

Bulk chemical and gas stores

Fire fighting pumphouse and reservoirs

For maximum safety benefit, where there are two mutually redundant sets of safety plant external to the main buildings these are generally sited well apart. Thus, for example, the auxiliary shutdown and diesel building, together with its associated essential aux­iliary transformers and fuel tanks, is situated about 140 m from the main control and essential diesel buildings and their auxiliary transformers and fuel tanks on the opposite side of the reactor building. Similarly, the station unit and auxiliary transformers are separated into two groups, one on either side of the turbine house. Groups of water storage tanks of similar duty are situated on opposite sides of the main buildings and remote from the towns water reservoirs. Cable routes from both the main control building and the secondary diesel and control building follow segregated routes. The same applies to cable routes between the station transformers and the 132 kV sub­station, and between the generator transformers and the 400 kV substation. Essential service water pipe­work from the CW pumphouse to the auxiliary build­ing and component cooling water pipework from the auxiliary building to the reserve ultimate heat sink is also segregated.

To provide assurance that these high integrity systems remain in an optimum working condition, relatively

frequent testing has to be carried out. The testing is very detailed and has to prove that every protection measurement is functioning correctly and is capable of providing trip output if required".

For the secondary protection system which employs analogue amplifiers and Laddie guard lines, test trol­lies are provided and the testing is done manually.

For the primary protection system, the use of computers has enabled a considerable proportion of the routine testing to be automated. The automated system provides printed records of the tests carried out.

Axial macroscopic

Axial macroscopic rating shape is basically a cosine shape, modified by the presence of partially-inserted control rods and the effects of burn-up. The burn­out of fission cross-section is the dominant cause of rating shape change in AGRs and leads to pronounced flattening. Figure 3.8 shows a typical rating shape in a new fuel stringer and in fuel irradiated to 24 GWd/t channel average irradiation. These rating shape changes with burn-up were calculated for a single channel surrounded by a sea of fuel representing fuel for an equilibrium fuel cycle, i. e., a mixture of irra­diations from zero to discharge. The curves show the characteristic ‘irradiation flattening’ produced by burn — up which results in a reduction in axial form factor,

channel discharge irradiation. These were calculated using a supercell calculation of an array of fuel channels at irradiations spread over the range from zero to discharge irradiation. Burnable poisons are incorporated in the 24 GWd/t fuel shown and it can be seen that the poison has, as well as restricting the initial reactivity as shown in Section 2.4.1 of this chapter, restricted the change in age factor over 24 GWd/t burn-up to the same range as for 18 GWd/t fuel. The age factor effect is an important contributor to the overall radial form factor change with burn — up, which is also affected by other macroscopic flux shape effects such as zone difference and control rod movements. In the initial fuel charge, the age factor starts of course at zero and the radial form factor is therefore small early in core life and is dominated by the macroscopic flux differences, which are mini­mised by the use of differential enrichment in two or three enrichment zones.

Rating fine structure

Burn-out effects also occur in rating fine structure distributions. These have significant impact on the

i. e., the ratio of peak to mean rating has reduced from 1.22 to 1.14. It can be seen that even the rating shape for a new fuel stringer is somewhat flattened because of the flux distribution formed by the sur­rounding fuel and the effect of the axial reflectors; a pure cosine shape would give a peak to mean rating ratio of 1.57.

The irradiation flattening clearly does not apply in the early stages of initial fuel charge. In order to flatten the axial form factor in initial charges, a non — uniform enrichment loading is used in which the top and bottom few elements are of a higher enrichment than the remainder.

Care is taken in reactor operation to ensure that:

• Certain operations are carried out in the correct

sequence,

• Appropriate protective devices are in service be­fore commencing operations which could introduce faults against which those devices are designed to

protect.

• Certain faults which could occur during start-up are prevented from occurring, if prevention is bet­ter or easier or cheaper than providing protection against them.

Precautions can be taken by means of administrative control such as mandatory instructions and checklists, but some are so important that it is deemed necessary to ensure them by means of interlocks.

An example of the first category is the sequencing of bulk rod withdrawal, i. e., in a four-group arrange­ment the bulk rods must be withdrawn in a sequence of groups 1 —2—3—4; electrical interlocks ensure that group 2 cannot commence withdrawal until group 1 is virtually fully withdrawn and so on. Such sequenc­ing interlocks are similar in principle to those ad­opted on fossil-fired plant, for example, on a coal pulverising mill group the mill cannot be started until the exhauster fan is running.

An example of the second category is loss of feed — water protection; this protection can safely be vetoed (rendered inoperative) while the reactor is shut down, but electrical interlocks may be used to ensure that the vetoes are removed (protection becomes operative) before reactor power rises above a few MW; the elec­trical interlocks may be arranged so that the reactor will trip if the vetoes are not removed.

An example of the third category is reactivity faults arising from erroneous regulating rod movements dur­ing start-up; it is difficult to provide adequate pro­tection against such faults, so interlocks are used to ensure that these rods are either fully withdrawn or in a given position in the core and on ‘hold’ supply before bulk rod withdrawal can commence.

Thus interlocks ensure that the correct procedures are followed; they do this either by preventing the start-up from proceeding unless the correct conditions are satisfied, or by causing a reactor trip if the cor­rect conditions are not satisfied.

Requirements and scope

The materials, components and structures of nuclear reactors have mechanical properties which are de­pendent on temperature and often a function of both temperature and time. An example is the creep of metals. Temperature measurements are therefore most important in providing the means to prevent tempera­tures reaching the point where the performance of vital parts of nuclear reactors is no longer satisfac­tory and in the limit fail with serious consequences to safety, for example, the maximum temperature of the fuel.

The various parts of nuclear reactor systems and other plant have different temperature coefficients of expansion and thermal capacities and so their per­formance is dependent on rates of change of tem­perature. As a result, the measurement of rates of change of temperature is important for some parts of nuclear reactors. An example is the reactor core, its restraint structure and steel/graphite samples, dis­cussed in Section 6.6.3 of this chapter.

Measuring techniques

Thermocouples are almost exclusively used for tem­perature measurements as described in Chapter 2.

The temperatures of the reactor core and structure are measured by thermocouples in direct thermal con­tact but it is possible to cover only a limited number of fuel elements in this way. The majority of the fuel element temperatures are inferred from the chan­nel gas outlet temperatures.

Gas temperatures at the exit and entry to the boilers are measured by arrays of thermocouples, and the mean temperature is calculated from a number of measuring points across the gas ducts.

6.3.1 Other measuring systems

Many other measuring systems are used to provide information on the state of the reactor. The more important ones are: [34]

tentiometers are used to make the measurement and the control rod position is displayed in the control room.

• Coolant pressure.

• Coolant composition.

• Water purity.

The total reactor power generated can be calculated from the product of the gas flow rate, specific heat and temperature rise over the core.

Normally the reactor power is displayed on the con­trol desk in two forms:

• Absolute power, derived from a thermal column adjacent to the core measuring the neutron density. Power is linearly related to neutron density in the

system.

• Backed-off linear power where the deviation from the normal steady state level is indicated.

Power assessment during normal operation

The true power level is that as calculated since the linear power indicator or recorder relies on seeing neutrons at one column only adjacent to the reactor core. The calibration is not automatically corrected if the reactor core is unbalanced in power generation. Movements of power in the reactor may occur over long periods, principally in the axial and radial planes. However, the linear power meter in association with the backed-off power meter gives a good indication of any immediate change in the reactor performance.

The daily calculation of power indicates the state of trim of the reactor and enables a comparison to be made with readings on the linear power meters. It is also a measure of the true power being pro­duced at the time of any reactor shutdown for the purposes of calculating afterheat (decay heat) and to give an estimate of the quantity of reactor poisons. The degree of xenon poisoning is necessary to deter­mine the control rod positioning at criticality.

Power assessment during shutdown

hen a reactor is shutdown, heat is being produced by the decay products from the fission process. This power, which commences at approximately 5wo of the pre-shutdown heat generation, decays exponentially and although heat generation falls rapidly it will be many weeks before it can be said to be insignificant.

The capacity of gas circulators and feed systems are designed to be adequate to handle the decay heat. However, an estimate at the heat removal rate is ne­cessary to determine the time required for components to reach acceptable temperature levels to enable main­tenance under depressurised conditions to commence.

An alternative design approach to thermal reactors is to dispense with a moderator and have the chain reaction maintained by fast neutrons (Section 9.1 of this chapter).

The attraction of the design lies in the characteris­tics that the average neutron yield for fission caused by high energy neutrons is greater than for thermal and that all absorption cross-sections are reduced. It is possible for ‘spare’ neutrons to be utilised in con­verting fertile to fissile material at a rate that exceeds the consumption of the fissile material required to sustain the chain reaction. That is, fuel ‘breeding’ is made possible and ultimately enables up to about fif­ty times more energy to be extracted from a given amount of initial fuel compared with its use in a ther­mal reactor.

A number of countries have been active in fast reactor research over many decades; the fast reactor has the distinction of having been the first to generate ‘nuclear’ electricity, in 1952, in the US Experimental Breeder Reactor (EBRI), Even so only one commercial size power station has been built, the 3000 MW(th) Superphenix by France. Other existing fast reactors must be regarded as prototypes and include the UK’s 600 MW(th) Prototype Fast Reactor (PFR) sited at Dounreay in Scotland. The reason for this lack of large scale exploitation is that the fast reactor system

quires a large initial capital expenditure and is justi­fied only if fuel costs is a maJ°r faccor in lhe COm’ Detine thermal sstcm.

Fast reactor design (Table 1.8 and Fig 1.33) must take into account that high fuel enrichment is neces — v (about 20°Т)) and that the absence of a moderator ojes a small core volume and hence very high power density of about 500 MW m- To emphasise this last point a 3000 MW(th) core needs only to be a 2 x 2 m diameter cylinder; the high heat extraction rate re­quires a liquid metal coolant (hence the term LMBFR), although high pressure helium has been proposed.

The world’s common approach to the above, is to use ceramic uranium fuel (UO2) ‘enriched’ with 20^0 added plutonium 239 as PuCh and to use liquid sodium for the coolant. The fuel-breeding (the con­version of U-238 into Pu-239) by neutron capture takes place in the U-238 blanket surrounding the core. This may be a natural or depleted uranium.

Sodium, Na-23, reacts readily with water; it also unfortunately captures neutrons to become the radio­nuclide Na-24, (3/y active with 15 hours half life. Consequently, sodium cooled fast reactor designs use an intermediate sodium/sodium heat exchanger to se­parate the radioactive primary sodium from the steam generator and to prevent boiler leaks causing water ingress to the core.

Figure 1.33 shows the pool type of design in which the core, primary pumps and intermediate heat ex­changers are all submerged in liquid sodium within a large essei. This is the design of UK Prototype Fast Reactor (PFRj and the French Superphenix. Some countries favour the alternative loop type, where the core only is submerged in sodium, the primary cir­culation pumps and heat exchangers being located externally.

Significant features in the overall safety of the fast reactor are that neither the pool nor the loop system is pressurised, and that the sodium coolant has a large heat capacity.

A commercial fast reactor core consists typically of 100 000 fuel pins, each pin being a stainless steel tube containing 5 mm diameter UO; pellets enriched in PuO;. Two to three hundred fuel pins encased in a stainless steel wrapper form a fuel sub-assembly. Re­fuelling of sub-assemblies at about 100 000 MWd/t is off-load from above the core. Control rods also enter the core from above and are generally boron carbide or tantalum in stainless steel.

During major reactor overhauls it is necessary for ‘man entry* to be made into the primary coolant cir-

0 OS 10 15 20

CARBON MONOXIDE. VO

Fig. 1.52 CAGR coolant window

cuit for component inspection and the atmosphere must be changed from carbon dioxide to air. As oxygen will react both with the graphite in the core and with uranium/uranium oxide (if failed fuel is present) a maximum temperature limit of 200°C is applied at which it is permitted to have air in the reactor. Imposing this maximum temperature ensures that reaction rates are reduced to negligible values.

The lowest temperature surfaces, in both magnox and AGR gas circuits can accumulate ammonium chloride and possibly iron chloride. In the presence of such soluble contaminants surface wetting occurs at a lower gas moisture concentration than for a clean surface, the point at which wetting occurs being ex­pressed as a percentage of the relative humidity (RH). Wetting of such salts can lead to enhanced corrosion and pitting of mild steel surfaces. Experiments have demonstrated that wetting of surfaces occurs at 50% RH for ammonium chloride contamination and 35% RH for iron chloride contamination. Irradiation of stainless steel fuel can material in the temperature range 350-500°C can lead to sensitisation and possi­ble subsequent corrosion when exposed to ‘damp air’. Experiments have been carried out which demonstrate that corrosion will not occur at 10% RH at the fuel can temperature. It is recommended therefore that during periods of an air environment the moisture level is kept both below 35% RH at the lowest metal temperature and 10% RH at the lowest fuel can temperature. The minimum fuel can temperature is higher than the lowest metal temperature such that the two limits give a maximum water-in-air concen­tration of 4000 vpm at atmospheric pressure.