[n the core design phase, features which determine life are sized on the basis of best available data. Much of this data uses results from small specimens irra­diated in uniform fluxes and temperatures. The major problem arises in applying such data to the larger geometries of graphite blocks, due to variable dose and temperature, together with problems of internal strains tempered by irradiation creep. Further, the variabi­lity of radiolytic loss produces property and strength changes throughout the brick. Complicated computer programs have been evolved to examine brick strength and weight loss, but these can only be validated against shorter-life simplified-geometry experiments. For these reasons, it is essential that the predictive design basis is both monitored and checked against new data, so that reactor life can be optimised to achieve highest economic benefit from the capital investment and on­going running costs.

Equipment that has been used in magnox and AGRs includes:

Interstitial channel samples This technique involves placing characterised graphite samples in interstitial channels either directly exposed to the coolant or con­tained within other graphite samples, to study depth effects. These can be removed periodically for mea­surement and, in some cases subsequently reinstalled. Parameters that are measured include weight change and deposit levels (giving radiolytic oxidation-rates), dimensional change, strength change, air thermal oxi­dation rates and stored energy. Whilst the accuracy of measurements from such pre-characterised samples is high, the technique suffers from the major dis­advantage that the dose received is only 50% of that received by the channel bore. Thus, in general, the measurements represent data confirmation rather than actual core state.

C~U monitor Most of the AGR reactors are pro — sided with C-14 capsules (see Section 10.4 of this chapter) which give an instantaneous measure of the graphite-oxidation rate. The equipment is fragile, par­ticularly the calorimeter electrical circuits, and it is tune-consuming to process the data. Nevertheless, it ls or*ly technique that can give an early indication °f core corrosion and has been used with success, to demonstrate that the core is corroding as predicted.

The channel bore monitoring unit (CBMU)/trepanning tool These two items of equipment may or may not be installed on the same carrier but, in any event, are used with the reactor shut down and depressurised. The CBMU gives channel profiles, i. e., integration of individual brick tilt, as well as diametral measure­ment, brick bow and channel discontinuities. In the later reactors all the channels were characterised by the CBMU prior to reactor start-up to give precise datum information. In the early designs of reactor, the pre-irradiation characteristics of individual bricks in known positions were not measured. As the graphite properties are variable from brick to brick and the changes in early life relatively small, the trepanning data obtained to date has a large uncertainty. When the importance of trepanning in giving definitive data on the state of the core was recognised the designs of reactor were modified to include up to 40 channels of individually characterised bricks.

PIE (post irradiation examination) of AGR fuel strin­ger graphite sleeves In AGRs, the graphite sleeves are integral with the fuel element and hence are re­placed at each refuelling. Measurements are made on both the whole sleeve and small specimens cut from the sleeve to give data on graphite oxidation rates, strength changes, dimensional and permeability changes and generally have confirmed predicted behaviour. An example of this is shown in Fig 1.50, which com­pares the predicted dimensional change with the mea­sured changes. The internal stresses generated by the thermal gradients and the fast neutron flux gradients are also measured. The technique involves machining an axial slit of known width in the irradiated sleeve. The internal stresses tend to cause the slit to close by a uniform amount over the main body of the sleeve from which the internal bend stress may be calculated. It has been shown that the closure and hence the internal stress of an ‘as machined’ sleeve is small.

Magnox core monitoring from station chemistry re­cords In magnox reactors the only source of carbon monoxide (apart from adventitious oil ingress) is from the radiolytic corrosion of the graphite core. The rate of carbon monoxide production over any given period may be calculated from:

• A knowledge of the change in carbon monoxide concentration in the reactor gas (coupled with the total mass of carbon dioxide).

• A knowledge of the quantity of carbon monoxide leaked from the circuit (based on leak rate mea­surements using helium as the tracer gas).

This method can give day-to-day mean core corro­sion rates and hence the effect of short term changes in the coolant composition can be determined. These

mean-core corrosion rate estimates provide a valuable cross-reference with the interstitial channel samples, the latter necessarily being cumulative over long peri­ods of reactor operation. This technique cannot be used in AGRs because the radiolytic oxidation of methane yields greater than 90% of the carbon mono­xide, with graphite corrosion only providing the re­maining less than 10%.

Defective fuel elements

Failures in the fuel element envelope are serious because this can lead to a deterioration in heat trans­mission to the coolant, in which case bar centre temperatures can exceed 1000°C during uranium oxi­dation. This raises the possibility of can ignition and core contamination by fission products released from the fuel.

Failures are detected by BCD monitoring which identifies certain short-lived gaseous fission products. This is discussed in Section 1.7 of this chapter.

Fast failures involve the following potential hazards:

• Oxide loss; a few grams of oxide can cause sig­nificant circuit contamination and a fast burst will typically contain 30 to 50 grams of oxide.

• Reduction of ba^ section; this can be serious in the stacked design of herringbone element.

• Overheating and ignition.

Only small numbers of fast bursts have occurred in CEGB reactors and none have caused reactor damage. They arise from pre-existing defects, typically damage to the end weld either during manufacture or pile cap handling. Less often they have arisen from defec­tive welding or defective end caps.

Examination during post-irradiation-examination revealed that some cans had a very coarse grain size and that failure occurred by cracking between grains of similar orientation.

General endurance studies

An important point to remember was the imposition of a 360°C gas outlet temperature limit in 1969 which reduced peak can temperatures in most CEGB reactors by 40° C.

In the lower part of the stack the main problems are associated with uranium growth, creep and swelling and with magnox ductility. Above half-stack height magnesium creep problems predominate.

Can torsion

It was observed that the fuel elements twisted (i. e., tightened the can helix) or untwisted (i. e., unwound the can helix) during irradiation with the torsion values reaching 80°. Examination revealed that these effects were caused by slight residual textures in the rim and core of the bar. This has not proved to be a serious problem.

Neutrons do not themselves cause ionisation and they are detected by means of ionising products of neutron-induced reactions on boron or uranium. The principal reactions used are the B10 (n-a) Li7 and the neutron fission reactions.

The thermal column graphite thermalises the in­cident neutron flux, the thermal neutrons producing an n-a reaction in boron-10. Slow neutrons may also be detected by the fission reaction with uranium-235 which yields heavily ionising fission fragments.

In detectors using boron in the form of BF3 most of the energy of the charged particles produced in —
the neutron reaction is absorbed by the gas with which the detector is filled, producing an ionising track. With the fission reaction less than one-third is absorbed.

The ionisation may be collected by applying an electric field by a polarising potential between the electrodes of the detector and in a neutron flux this flow of charge is measured either as a mean current or as a succession of ionising events or ‘pulses’. De­tectors intended for operation in the former manner are known as DC (mean current) ionisation chambers and those operating in the latter manner as pulse counters. A method known as ‘Campbelling’, described in Section 5.2.12 of this chapter, provides a detection system intermediate between counter and mean current, and operates over a wide range.

A general arrangement of the Stage 2 fuel element is shown in Fig 2.73. Features to note in comparison to the Stage 1 fuel are the single thick stronger graphite sleeve, the integrally welded sleeve/grid/brace struc­ture, and profiling on the the outer surface of the graphite sleeve to reduce stringer vibration during charge/discharge operations.

The grids and braces are retained within recesses in the graphite sleeve bore by means of a composite assembly spot welded in situ.

This fuel type has been adopted for the first charges of Heysham 2/Torness reactors and will form the basic feed fuel for all AGRs from 1987 onwards.

The beneficial strength of the sleeve support struc­ture and performance advantages of the streamlined brace and variable channel bore size have been de­monstrated in out-of-pile tests already confirmed by reactor pilot loadings,

6.1.1 Fuel element performance

An ongoing programme of post irradiation examina­tion of discharged fuel has confirmed the capability of the AGR fuel and given reassurance that the target burn-up of 18 GWd/t can be increased to 24 GWd/t with full confidence.

The essential electrical system, post-trip sequencing equipment and mechanical plant in the quadrant areas is segregated (by barriers or distance) on the basis of the four trains, with the X and Y systems to­gether. In the case of the common mechanical sys­tems outside of the quadrant areas this is impracti­cal; in this instance the X plant is segregated from the diverse Y plant, and within particular systems further segregation is provided between pairs of trains (AB and CD). This arrangement prevents any single major hazard from disabling all post-trip heat removal systems.

Note:

1 Numbers represent the number of quadrants required to establish successful post trip heat removal. 2. The gas circulators are required to operate in the same quadrant as the associated boilers

Fig. 2.113 Post-trip heat removal systems

8.1.2 Automatic sequencing of plant

The post-trip heat removal systems are brought into service automatically by the actions of the post-trip sequencing equipment; this is designed to achieve satisfactory reactor cooling under all fault conditions without assistance from the operators within the first 30 minutes following the reactor trip. In order to simplify the system, the same basic sequence (with some minor modifications) is followed for all fault conditions; plant status signals are not required to establish this basic sequence.

The reactor pressure vessel, illustrated in Fig 2.129, may be considered as the central component of the reactor coolant system. Its purpose is:

• To support and to keep in position the reactor core, the internal structures and the control rod

drive mechanisms.

• To contain reactor coolant within a leaktight bound­ary having a high resistance to failure under inter­nal pressure. [23]

The vessel consists of a flanged vessel shell with in­let and outlet nozzles, and a flanged removable up­per closure head with control rod drive mechanism (CRDM) housing adaptors.

The vessel shell consists of a steel cylinder, mounted with its axis vertical and closed at its lower end by an integral domed head. The vessel shell is constructed of ring forgings with a single piece dome forging for the lower head. The cylindrical section comprises a lower, plain portion approximately 216 mm in thick­ness and 4,394 m internal diameter, surmounted by a region of thickness approximately 267 mm containing four inlet and four outlet nozzles, of internal dia­meter 698 mm and 736 mm respectively. The nozzles are arranged symmetrically around the vessel and con­nect the vessel to the steam generators and reactor coolant pumps via the primary coolant piping. Above the nozzle course is a flange containing threaded holes for bolting the closure head. The internal diameter of the flange is machined to provide a ledge from which the core and vessel internals are supported. The pressure vessel shell is designed so that changes of section, structural discontinuities and stress concen­trations are outside the region subject to high neutron irradiation.

At the lower end of the vessel cylinder, just above the lower head, guide pads on the inner surface of the vessel provide lateral location and restraint for the lower end of the reactor internal structures. The bottom head of the vessel contains penetration nozzles for connection to and entry of the nuclear in-core instrumentation.

The removable upper closure head consists of a hemispherical dome with a bolting flange at its peri­phery. The flange contains two flat-bottomed circum­ferential grooves which house hollow metallic O-rings to seal the joint between vessel and head.

Seal leakage is detected by means of two leak-off connections, one between the inner and outer O-rings and one outside the outer ring. The closure head is provided with lifting lugs to facilitate removal for re­fuelling and for access to the internal surfaces of the vessel for in-service inspection. The dome portion of the head is penetrated by a vent line near its centre and by adaptors for the control rod drive mechanisms and above-core instrumentation.

The vessel assembly is supported from pads located on four of the main coolant nozzles (two inlet and two outlet). The supports are spaced uniformly around the vessel at 90° intervals.

At the design stage, the vessel is subject to compre­hensive stress analysis to ensure that ample margins exist against failure under all conceivable operating conditions. The reactor vessel is manufactured to very high standards of quality and is subject to repeated, thorough inspections to ensure that it enters service free of defects which could result in structural failure. During service the condition of the vessel is moni­tored by further inspections to confirm the absence
of defects, and by tests on material samples to con­firm that the properties of the vessel material have not deteriorated unacceptably.

1 Operation policy

1.1 Licence requirements

Before any nuclear installation or reactor can be built or operated a licence must be obtained. The licence is issued by the Secretary of State for Trade and Industry through the Health and Safety Executive (HSE). The principal statutory requirement governing the operation and construction of nuclear power sta­tions is the Nuclear Installations Act 1965 and 1969, although a number of other acts also placed statutory requirements on the CEGB (and its nuclear successor, Nuclear Electric pic) when operating its installations. A brief description of these principal requirements is given under the following headings.

1.1.1 Nuclear Installations Act 1965 and 1969

To illustrate the requirements of this Act, direct quo­tations are given on its main provisions:

• ‘By licensing, to control, in the interests of safety, the building and operation of nuclear reactors and similar installations and nuclear matter in transit.’

The site licence must be obtained before any instal­lation or operation of any plant takes place for the production or use of atomic energy. Similarly any processes used for the production of atomic energy or for storing or processing nuclear matter (including fuel), must have a licence. The extraction of pluto­nium from irradiated fuel or the enrichment by in­creasing the proportion of Uranium 235 is forbidden on any site.

Copies of the site licence or its conditions have to be posted upon the[26] site according to the directions of a site inspector of the Health and Safety Executive (HSE).

The duties of the licensee are to ensure that in pursuance of generating nuclear energy no occurrence on the site will injure any person, other than the li­censee, by way of ionising radiation. Similarly he is responsible for any nuclear matter which is in transit from the site, and this would normally include nuclear active waste for disposal or irradiated fuel being dis­patched to British Nuclear Fuels for processing.

As well as providing financial cover for the licen­see’s liability, the Act does make provisions for time scales for claims or compensation. This is within 30 years from the date of the incident or 20 years if the licensee has lost possession of nuclear matter by way of theft, loss, jettisoned or abandoned. After this period any claim is made to the Secretary of State for Trade and Industry.

Dangerous occurrences of a type prescribed by the Secretary of State are reportable to him. He has made the Nuclear Installations (Dangerous Occurrences) Re — eulations which prescribe those occurrences which are reportable to him, In the event he may order a special inquiry which may be public, or direct an inspector of the Health and Safety Executive to make a special

report.

Qualified inspectors appointed by the HSE can enter any site or premises for the purpose of any test and/ or inspection considered necessary. The inspector can require the licensee or anyone with duties connected with the licensed site to provide such information or permit inspection of documents as he may require.

There are three types of feedback mechanisms which occur in gas-cooled reactors. They may be classified as those which arise from:

• U-238 resonance effects.

• Neutron spectrum changes which influence the be­haviour of other nuclear isotopes.

• Dimensional changes.

In addition, in water reactors there are feedback ef­fects due to changes in density of the coolant.

3.1.1 U-238 resonance effects

When a neutron is absorbed by a nucleus the energy of the nucleus increases by the kinetic energy plus the binding energy of the neutron. If this total energy is near one of the energy levels of the compound nucleus, the probability of formation of the com­pound nucleus increases dramatically. This nucleus may decay in several ways:

(a) The neutron may be re-emitted with or without gamma-rays (inelastic or elastic resonance scat­tering).

(b) The nucleus may emit gamma-rays (radiative cap­ture).

(c) The nucleus may divide by the process of fission.

Option (c) is not possible for U-238 at low neutron energies. Options (a) and (b) compete as modes of de-excitation, with (a) having a low probability at low energies because the neutron has to gain sufficient enerey from the other nucleons to overcome its bind­ing energy, and (b) having a high probability. Ra­diative capture is a relatively slow process on the time scale of nuclear processes (10’14 seconds) as witnessed by the very narrow peaks of the resonance in the energy range 6.67 ev to 200 ev. Neutron re­emission is a fast process at high energies U0_:i seconds) and the resonances are broad.

The widths of the radiatіe capture resonances are temperature dependent. When the target nucleus is vibrating thermally the velocity of the neutron relame to the nucleus is spread over a range of velocities. The effective capture cross-section away from the peak of the resonance is, therefore, increased with increasing target thermal energy. By an analogy with the effect of a moving source or observer in the theory of sound waves, this broadening of the radiative capture cross — section is known as Doppler broadening. It can be shown that as the width increases so the height of the resonance decreases and the total probability of capture within the resonance remains constant. The temperature effect arises from self-shielding in a he­terogeneous reactor. In general, the peaks of the reso­nance are so high that all absorptions occur on or near the surface of the fuel. With a broadened reso­nance the neutrons near the peak of the resonance are still absorbed, albeit slightly further into the fuel, and neutrons in the ‘wings’ of the resonance are absorbed in addition. Consequently as the temperature of the fuel rises so the number of absorptions in U-238 in­creases giving a negative temperature coefficient of reactivity.

Plutonium isotopes produced during irradiation also show resonances but they do not contribute to tem­perature feedback as they are produced mainly on the surface of the fuel as a result of U-238 capture — there is very little self-shielding in this case (PWR is an exception to this — the very high burn-up of PWR fuel results in larger amounts of plutonium isotopes being built up and the self-shielding mechanism starts to come into play).

A reactor trip may be initiated automatically by the safety systems when a fault condition or plant ab­normality is detected. It may be a problem on the reactor itself, or a problem on the boiler, main tur­bine or boiler feed system which requires a prompt reactor trip. These latter points are applicable par­ticularly to AGRs, where the unit arrangement of reactor and turbine, the high power density of the system, the small water capacity (and hence small heat capacity) of once-through boilers and the need to closely control the conditions in the boilers be­cause of their materials and construction, require that automatic systems be used where prompt action is necessary.

The reactor trip may occur because of a fault in the safety systems themselves, when the plant is healthy, but this is an unusual occurrence assured by careful design of the safety systems.

A reactor trip may also be initiated by the reactor — control engineer. This may be a situation in which a rapid shutdown is required, for example, a major problem in the turbine hall requiring the turbines to be taken off-load quickly. It may be a fault condition which should have been dealt with by the automatic safety systems, and the Operating Rules require that if the operating staff perceive a situation which should have shut down the reactor automatically, they should do so manually. It may be a fault condition which is not protected against by automatic safety systems but relies on operator action, for example, a sub­stantial increase in fission product activity indicated on the burst can detection (BCD) equipment. The reactor control engineer may also initiate a trip as part of the controlled shutdown procedure.

When the reactor is tripped from power, all control rods are released into the reactor core.

When an AGR is tripped a number of other actions are initiated, for example, the associated main turbine is tripped, the boiler feedwater is shut off (to avoid overfeeding the once-through boilers) and the gas cir­culators are tripped (to avoid undesirable temperature transients in the below-core region). These and other

important actions are carried out in the desired sequence and at the appropriate times by sequencing equipment dedicated to this purpose. This sequencing equipment also starts up the required plant for post — trip cooling. The only immediate actions required of the reactor control engineer are to check that the sequence is carried out correctly.

Scope

Measurement of coolant flow is relevant to operation because the power output of the reactor is related to ic The coolant is pumped through the core by gas circulators, but because of the complex geometry, direct flow measurement by conventional methods is not possible.

Coolant flow may be derived in one or both of the following ways:

• Measured in gas ducts using a form of pitot tube (at some magnox stations with steel pressure vessels, e. g., Berkeley, Bradwell).

• Calculated from circulator characteristics, pressure and temperature (all stations).

Limits of coolant flow are set in operating rules to:

• Protect circulators from excess flow problems, e. g., vibration.

• Limit channel/reactor power, the reactor operation not being normally related to a direct power limit.

Circulator speed is used to calculate coolant gas flow, circulator speed interlocks, etc. Failure of speed indi­cations has led to loss of circulators and hence forced cooling tn some incidents.

Circulator instrumentation is provided to ensure that the machines operate within their design envelope with healthy margins, during normal operation and post-trip. In particular, to provide a monitoring capa­bility to give a high reliability of a sufficient number of circulators operating post-trip to assist decay heat removal.