The most common fuel failure mode in the UK mag- uo reactors has been body failure. This occurs near i-e position ot peak can temperature, the strains im — po^’d on the can by uranium swelling causing it to lail by creep cavitation at subgrain boundaries within — uee grains. Analysis of the failure frequency has in­dicated a strong correlation with element manufac­turing number (which is, in turn, probably indica­tive of minor variations in the heat treatment); thus, once failures begin to occur, other fuel at risk can be quickly identified and remedial action taken. As in other cases, this provides a lesson in the usefulness of complete historical records for each fuel element. Currently the failure rate is about one element per reactor year.

Another cause of fuel failure is coolant leakage through damaged or faulty end cap welds. Since cool­ant ingress to the fuel occurs as soon as the reactor is pressurised, this leads to progressive oxidation of uranium at a position remote from the weld with, at first, very little release of activity. Subsequently, the expansive forces generated by the uranium oxide cause

the can to fail and produce a high-burst can signal. However, such ‘fast-bursts’ have been shown to be less serious than was originally supposed since, even when oxidised right through, a failed fuel element is still capable of supporting the weight of the fuel stack above it. In addition, such failures are generally li­mited to low burn-up fuel since, at higher irradia­tions, creep down of the clad onto the fuel prevents extensive in-leakage of coolant gas.

As explained previously, the generation of inter­nal stresses in the uranium bar by irradiation growth leads directly to yielding creep. Because this is a low temperature phenomenon it can lead to bowing of the bottom elements of the fuel stack, where the temperatures are lowest and the axial loads highest. The elements bow until the splitters or lugs touch the channel wall, after which the bow shape becomes progressively more complex. However, little channel distortion has been found and, since it is limited to low rated positions, the heat transfer penalties of such bowing are minimal.

Magnox AL80 oxidises only slowly in CO2 forming a strong, adherent film. However, at the higher clad temperatures, the presence of certain impurities in the coolant — notably water or the higher hydrocarbons — can cause the film to produce tensile strains in the fins, especially the thinner sections. Since the fins taper towards their tips, the net effect is that they become wavy with considerable loss of heat transfer capability. Such strains are most severe in the thinner — finned polyzonal elements. In the UK this problem arose at about the same time that severe corrosion of

jld steel components was found in the reactors. This, ner problem led to an upper limit of 360°C being nut upon channel gas outlet temperature and a side effect of this was that it effectively eliminated tin

w a’dng.

The main primary coolant circuit construction ma­terials, together with a typical hard facing material and fuel clad are identified in Table 1.20 in terms of і heir elemental composition. It should be noted that other elements will be present at trace levels, and that not all the materials in Table 1.20 are necessarily present in all PWR designs. For example, for steam generator tubing Incoloy 800 is widely used in Euro­pean PWRs whereas American designs currently use inconel 600.

These materials will always be covered with an initial oxide film, which develops and may undergo chemical modification due to prolonged exposure to the primary coolant water chemistry over the tempera­ture range of ambient to 300°C. Typically the oxide film on a stainless steel would increase from 1-2 fim to — ~ -1b Mm over 30 years. In general, such oxide films give protection to the metal surfaces against dgnilicant corrosion and loss of metal thickness, how — aer> ^1еУ can lead to contamination of the primary coolant by two mechanisms:

tow (up to 10 ppb), the nature of the species and their activation by passage through the core can lead to significant primary coolant activity levels.

• In addition, corrosion products can be released as discrete insoluble particulate material, containing insoluble and sparingly soluble elements. Such ma­terial is transported around the primary circuit and can deposit in the core, activate, re-release and re­deposit on out-of-core surfaces.

These two processes will be largely responsible for the circulating and deposited primary circuit activity levels, and hence plant radiation levels by a transport mechanism as summarised in Fig 1.54. Species re­leased initially from in-core surfaces will already be activated, whereas out-of-core material will become activated by residence in the core.

The extent to which metal species are transferred from circuit materials to the primary coolant will be a function of a number of inter-related factors, which determine the corrosion rate and corrosion product release rate. The corrosion rate will be largely deter­mined by the temperature, chemistry and any pretreat­ment of metal surfaces, whereas the corrosion product release rate will be a more complex function of these factors, including hydrodynamic and erosion effects. It should be noted that the chemical composition of released material will not necessarily reflect the com­position of the base material. For example, nickel release is often reduced and iron release enhanced with respect to the base metal composition. In addi­tion, there will be a contribution from wear processes in moving components such as valve internals and components of the control rod drive mechanism.

In order to relate the circulating and deposited levels of activity to the system corrosion and corrosion product release-rates, it is necessary initially to esti­mate the amount of corrosion products released by means of surface areas. In Table 1.21 system materials are listed against typical applications, corrosion and corrosion product release-rates and surface areas for a four loop PWR. The surface areas can be sub­divided in terms of in-flux, out-of-flux, and high temperature/low temperature wetted fractions, if there is a sufficient data base to permit refinement of the estimate of the amount of material released to the system. It should also be recognised that corrosion and release data can be an order of magnitude greater for the first few months of operation, and that sig­nificant increases can result from system transients (e. g., temperature, flowrate and pH).

The concentration and nature of circulating and deposited activated species has been the subject of much investigation, both to correlate with corrosion product release data and to optimise methods of re­moval. For a typical PWR at full pow’er the metal ion concentrations (soluble plus particulate) would be at sub-ppm levels and have been reported as iron

Table і.20 Composition of materials in contact with the primary coolant Cr Ni Fe Mn Sr C Mo W Co Nb Zr Sn 316 Stainless steel 16-18 11-14 Balance <£2.0 2-3 304 Stainless steel 18-20 8-11 Balance 0.07 Inconel 600 14-17 >72 6-Ю <£1.0 lncaloy 800 19-23 30-35 Balance <1.5 Inconel 690 27-31 >58 7-11 <0.5 1 Inconel 718 19 52 18 3 5 Zircaloy 4 0.1 <0.5 0.2 і Balance 1.5 Stellite 26.5 -30 0-3 0-3 0-0.5 0.7 — 1.5 0.9 — 1.4 0-1 3.5 -5.5 Balance

(10-20 ppb), nickel (2.3-3 ppb), cobalt (0.2 ppb), chromium (0.8-2.3 ppb) and manganese (0.3-0.9 ppb). A consideration of such data, together with release rates and surface areas, leads to the conclusion that 2-20 g of material is present in the coolant at any given time. It is apparent also that up to 25-30 kg of material can be released to the coolant per year, which can deposit on circuit surfaces or be removed by the CVCS clean-up system.

The insoluble component of circulating and de­posited material is generally referred to as ‘crud’, and post irradiation examination (PIE) is used to both remove and identify the nature of crud deposited on fuel clad surfaces (fuel crud).

In discussing the nature of the species present it is necessary to differentiate initially between soluble and insoluble. Conventionally, soluble material is de­fined as that which passes through a 0.45 /tm filter,
accepting that some particulates will therefore be classified as soluble. The chemical identity of metal cations present can be inferred from the composition of the circuit materials and corrosion products. Table 1.22 gives the principal nuclides together with their formation reactions and half-lives. This data could be expanded to include Mn-56, Mn-99, Te-99, Sb-122, Sb-124, Zn-65 and Nb-95, which derive from minor constituents, impurities and welding consumables.

As Ni-58 (67.9% natural abundance) is a signifi­cant proportion of the metal content of released cor­rosion products, Co-58 production is high. With a half-life of 72 days it will equilibriate in one reactor cycle and dominate the plant radiation levels during early life.

Co-60 is produced from Ni-60 (26.2%) and Co-59 (100%), where the latter represents a trace impurity in construction materials and a major constituent of

some hard facing alloys. Typical cobalt levels would be 0.1-0.2% in stainless steels and 0.01-0.02% in nickel alloys, but 60% in some hard facing alloys. The Co therefore has a lower activation rate than Ni, however, the 5.2 у half-life results in continued increase over many years. Although the rate of in­crease will depend upon the Co-59 trace impurity level, corrosion/wear of hard facing alloys, rate of release to the circuit and residence time in the core, ultimately it is possible for the Co-60 level to dominate the ra­diation level.

If a single pressure steam cycle was used for magnox reactors, poor overall efficiencies and net power out­puts would result. Figure 2.16 shows a typical heat diagram for the single cycle w’here it is apparent that the attainable steam pressure is dictated by the minimum temperature difference between the gas and saturated water at the ‘pinch point’. Attempts to im­prove the efficiency by introducing feed heating are offset by the necessary reduction in cycle pressure, as shown by the dotted line. In order to maintain the same cycle pressure and obtain the benefits of feed heating it would be necessary to increase the reactor gas inlet temperature, i. e., raise the gas line. However, blower power is very sensitive to this tem­perature and the effect would be to reduce the net electrical output from the plant.

Dual pressure cycle

The disadvantages of the single pressure cycle can be partially overcome by having a dual pressure cycle. A proportion of the steam is raised in a high pres­sure section which is in series on the gas side (as shown in Fig 2.17) with a low pressure section, which effectively cools the gas leaving the boiler to tem­peratures that enable blower power to be minimised. Increased complexity and capital cost limit a practical arrangement to two pressures.

The temperature heat diagram (Fig 2.18), shows how a higher operating pressure is possible for part of the cycle. As cycle efficiency is approximately proportional to the area under the water/steam line, and a higher feed temperature is possible, the overall thermal efficiency is much increased without a loss in net electrical output. For example, with a dual pres­sure system at 14 bar and 56 bar instead of a single pressure system at 14 bar, and w-ith a boiler gas inlet temperature of 385°C and a gas outlet temperature of 172°C, the net electrical output is increased from

Ї L5 Ct О 5-" u

Ї о

о о и

HEAT CONTENT

Fig. 2.16 Typical single cycle heat diagram

Fig. 2.17 Dual pressure steam cycle

62 MW to 69 MW whilst the thermal efficiency in­creases from 25% to 28%. The proportion of HP (o LP steam flow is typically 60% HP and 40% LP with pressure ratios of around 4:1.

There are several thousand fuel channels in the mag — nox reactor through which the coolant passes.

Ideally the maximum fuel surface and gas outlet temperature of each channel should be checked, but in practice only a proportion of the channels are monitored in magnox reactors.

The chromel/alumel thermocouple with stainless steel sheath is used for metallurgical and nuclear reasons in power reactors up to about 850°C (Volume F, Chapter 4). Above 850°C, the thermocouple may be affected by deterioration of the protective oxide film.

On magnox reactors the fuel element surface tem­perature is measured by attaching the thermocouple to the fuel element can. The finning of the can is drilled at a selected point to take the thermocouple hot junction. Both time response and the accuracy are affected by the integrity of the thermocouple fit. In some cases continuous ‘trailing lead’ thermocouples are used, but because of the complication in fuel­handling operations, channel multiple plug and sockets are widely used. Great care must be taken to ensure that the materials and the design of the plugs and sockets will maintain a high insulation value in the reactor environment.

The 3 mm diameter thermocouple, used on magnox fuel elements, has a resistance of about 2 П/m and the fairly high lead resistance together with the com­pensating cable results in a total resistance per ther­mocouple of the order of 600 fi. The output is about 40 per °С so that 16 mV may be delivered around normal power conditions in a magnox reactor. Some thermocouples are used for excess temperature pro­tection and others for measurement only.

For maximum reactor power, the fuel elements are operated at as high a temperature as allowed by fault considerations which take account of a margin above

operating temperature. When this margin becomes zero, a reactor shutdown is initiated via temperature trip amplifiers and the reactor safety system.

5.3.2 Other temperature measurements

Temperature measurements on magnox reactors are discussed in Chapter 3 of this volume.

The drier plant is provided for the continuous re­moval of moisture from the coolant gas. The incoming gas passes first through a recuperative heat exchanger where it is cooled by gas returning to the reactor. A water cooler further reduces the gas temperature before it enters one of the two absorption towers filled with desiccant. The desiccant is a bead form of silica gel which does not deteriorate significantly due to the trace organics which are in the coolant gas. The dried gas returns to the bypass circuit via the recuperative heat exchanger. After a period, the de­siccant becomes saturated and the gas flow is auto­matically switched to the other absorption tower. The desiccant is reactivated by part of the cooled gas from the recuperative heat exchanger which is mixed with some of the inlet gas to give a temperature of about 200°C. The gas from the absorption tower during the reactivation cycle passes through a water cooled intercooler before entering the separator vessel where the water is removed. The gas is returned to the gas flow through the absorbing tower in service. The extracted water contains tritium and care has to be taken in its handling and storage.

Recombination unit

The function of the recombination unit is to control the carbon monoxide level in the coolant gas. Part of the bypass flow passes through a platinum catalyst in the unit where the carbon monoxide is combined with oxygen. The catalyst is supported on a perforated plate surmounted by stainless steel mesh to prevent the carry-over of dust from the catalyst bed. Any hydrogen which is present in the gas is converted to water. The rise in temperature caused by the reactions is approximately 35°C.

The catalyst and the desiccant in the driers have a finite life and provision is made for their replacement

and storage.

Filters

The filters in the bypass circuit are designed to re­move particulate matter down to one micron with a high efficiency. In each filter there are a number of elements which are made from either sintered stain­less steel mesh or particles. An alternative design of element is made from packs of etched stainless steel discs. The filters can be back-blasted with clean car­bon dioxide to remove particulate matter and provi­sion is made for its removal and collection.

Electrolysis plant

The electrolysis plant provides oxygen for the recom­bination unit, hydrogen for generator cooling and for the methanation plant if necessary. Oxygen and hy­drogen gases from the electrolytic cells are collected and passed to their associated low pressure gas hol­ders. Compressors take the gases from the holders and their pressures are raised to values suitable for their safe and economic storage and for injection into the reactor system. Driers and sometimes water separators are installed after the compressors to ensure that the moisture content of the gases is within the specified limit. Gas cylinders are used for high pressure storage and may be supplemented with off-site produced gas. The storage capacity for these gases is relatively small, being equivalent to a few days’ usage.

8.1 Control and instrumentation — extent and differences from magnox reactors

The basic requirements of control and instrumentation on AGRs are similar to those for magnox reactors, though the higher operating temperatures, coolant pressure and greater power density place great empha­sis on temperature constraints in the reactor structure including the boilers, and on the detailed monitoring of reactor coolant composition and the feedwater steam circuit.

Because of the smaller margins and complex inter­relations, the number of measurement-, is larger and there is considerable emphasis on reliability since mam measurements are involved in safetv-related operations and the Operating Rules.

The C and 1 provisions on Hevsham 2 are tvpical of other AGRs and are based on the centralised con­trol and monitoring philosophy. Under normal op­eration each unit can be operated by a single unit operator in the Central Control Room (CCR) with a minimum of local-to-plant staff.

The extent of instrumentation on Heysham 2, com­pared with previous AGRs, has been increased con­siderably mainly because of an increase in the amount of auxiliary plant associated with post-trip heat re­moval. Because of the segregated train policy and diverse cooling requirements, this has further increased the amount of C and I equipment. Extensive use is made of computer-based control, display and monitor­ing systems.

Modulating control is provided to allow the unit performance to be optimised and to keep plant para­meters within working limits without requiring continu­ous operator supervision. Sequence control is provided to assist the operator in the start-up and shutdown of some plant systems. Both of these are provided by computer-based systems.

Automatic protection systems are provided to ter­minate all fault sequences for which prompt action is required with high reliability. Automatic post-trip sequence equipment is described in Section 8 of this chapter. These systems are independent of the main computer systems.

The process instrumentation required for reactor con­trol and protection includes those devices and their interconnections which measure temperature, pressure, fluid flow and fluid level in tanks vessels or pipework.

Tappings are provide for each control and protec­tion measurement on the NSSS. One instrument tap­ping of a narrow range level measurement may be shared by a pressure measurement or by a wide range level measurement, provided that the instruments con­cerned are associated with the same electrical separa­tion group.

The pressuriser water level instrumentation measures water level using differential pressure. The pressuriser has an upper and lower tapping, the upper tapping maintains a reference leg full of water by condensation of steam at the top. The transducer is connected with one side to the lower tapping of the pressuriser and the other side to the bottom of the reference leg. ^ater level is determined by the difference in pressure (between the pressuriser). The reference leg is lagged to prevent errors due to rapid changes of temperature.

There are four pairs of tappings on the pressuriser, giving four separate channels of level measurement for the PPS and station automatic control system.

The lower tapping in each measurement of pressuriser level is also used for a pressuriser pressure input to the PPS/station automatic control system.

Each steam generator has four PPS narrow range (NR) water level measurements using the differential pressure method described above for the pressuriser, with four separate pairs of tappings, one pair per separation group.

For use by the secondary protection system (SPS) each steam generator has four separate narrow range water level measurements from four pairs of separate tappings at the same levels as those of the PPS, one per separation group. Each SPS measurement channel employs a sealed reference leg with hydraulic isola­tors between the reference leg and the steam generator. The transducer is connected with one side to the lower tapping of the steam generator and the other side to the sealed reference leg. The measurement principle is identical to that of the pressuriser but without the use of a condensation pot.

For the PPS, each steam generator has one mea­surement of wide range (WR) level. The level mea­surements are derived by differential pressure in the same way as the method used for the pressuriser. The wide range measurement uses the same reference leg as one narrow range PPS measurement on each steam generator and therefore has a common reference leg tapping to the transducer, but a separate lower tap­ping to the steam generator is provided.

There are thus a total of 17 tappings to each steam generator as follows:

4 PPS reference leg connections 4 SPS reference leg connections 4 PPS narrow range connections 4 SPS narrow range connections 1 PPS wide range connection

For the RCS pressure, two separate tappings (one for the PPS, the other for the SPS) are provided at each of four branch line locations:

Loop 1 RHR suction line Loop 2 safety injection line Loop 3 safety injection line Loop 4 RHR suction line

The SPS tapping in each case is the tapping nearer to the main reactor coolant loop pipework.

For the measurement of reactor coolant flow, each coolant loop crossover leg has four tappings on the inside and four tappings on the outside of the coolant pipework elbow-. Flow is measured by differential pressure between correspondingly offset tappings on the inside and outside of the elbow, one pair for each PPS channel.

Each reactor coolant cold leg provides PPS coolant temperature measurement by means of fast-response resistance temperature detectors (RTD) mounted in appropriate pockets in the cold leg piping, and SPS measurements by means of thermocouples. The cold leg temperature sensors are located between the RCP outlet nozzle and the primary shield wall. There are 10 penetrations on each RCS cold leg allocated as follows:

— NR temperature — RTD

— NR temperature — RTD

— WR temperature — RTD

— NR temperature — Thermocouple

— NR temperature — Thermocouple spare

— NR/WR temperature — RTD spare

Each reactor coolant hot leg has a pocket with an RTD to provide a signal to the PPS guard line of the corresponding coolant loop. An installed spare hot leg RTD is provided in each loop.

The site licence requires the licensee to produce a maintenance schedule, which is submitted to the Nu­clear Installations Inspectorate for approval. The schedule consists of a number of items of plant and equipment which are directly connected with the sate operation of reactor systems, and which are in need of a regular system of maintenance at a prescribed interval. The items listed are those which if not func — tioning correctly could contribute to an unsafe con­dition during reactor operation:

• Testing of reactor guard lines.

• Maintenance of emergency boiler feed pumps,

• Maintenance of valves in the boiler feed mains.

• Testing of reactor vessel safety valves.

• Maintenance of burst cartridge detection gear, etc.

The schedule will show the prescribed frequency which will be at least two years so that it appears between the biennial overhauls. Some of the items may only be carried out on a complete shut down of the re­actor with the reactor pressure vessel at atmosphe­ric pressure, and therefore may only be done during the biennial overhaul, e. g., CCb make-up isolating valves.

Measurements of reactivity coefficients in magnox reactors and in the Windscale AGR have been carried out using a steady state technique, usually as part of a reactor start-up. The reactor power is increased in steps, allowing time for the temperatures to equi­librium at each stage. The change of reactivity is assessed from the movement of the (previously cali­brated) control rod group used to induce the power change. The temperature change is derived from in­stalled instrumentation. The technique has many lim­itations. It requires the use of a calibrated control rod bank. Since the reactivity insertion rate is depend­ent on the axial flux shape, the calibration should be carried out with an appropriate flux shape which may not be easy to arrange. The measurements of temperature become more difficult to interpret as the power level is increased; at low powers the reactor is essentially isothermal and mean temperature changes are easily related to measurement, at higher powers this is no longer true. The technique only gives an overall reactivity coefficient for the reactor, i. e., a combined fuel and moderator coefficient. In order to derive the moderator coefficient, the effect of the fuel coefficient must be calculated and subtracted. Whilst this is satisfactory for magnox reactors where the fuel coefficient changes only slowly with irradia­tion, the technique is not appropriate to AGRs as both fuel and moderator coefficients are strong functions of irradiation.

Because of the importance of the fuel component of reactivity feedback in the safety of the AGR, a new transient technique was developed which relied on the differences between the time constants of the various components to differentiate between their con­tributions to feedback. The method uses small per­turbations around a steady state to measure the fuel feedback coefficient directly. Calculated moderator coefficients are used as correction factors in the ana­lysis route, but the sensitivity to errors in these values is small.

In the measurement, the reactor is perturbed from a steady state by withdrawing a bank of control rods for a short time, sufficient to insert about 50 mN of reactivity, typically 10-20 seconds. The rods are then held in this position for 30-50 seconds then reinserted to their original positions. It is important that any auto-control loops which would affect reactivity in any way (e. g., reactor inlet temperature, grey rod position) be set to manual during the test. The result­ant flux transient is analysed using a specially written computer program APRECOT [2] to give the fuel coefficient. Such a test gives two estimates of the feedback coefficient, one after the rods are withdrawn and one after they are reinserted. Any disagreement between these values is indicative of errors in the moderator feedback correction factors.

The technique has been used extensively at Hinkley Point В power station to measure the fuel coefficient as a function of core irradiation. The good agreement between measurement and theoretical prediction has allowed the uncertainty allowances placed on the feed­back coefficient in safety calculations to be reduced substantially.

temperature

Bv reference to the formula given at the beginning of this Section 5.5, it can be seen that a change in oas flow will give rise to a change in power if tern* perature is maintained constant. This is the situation in laree power changes, as mentioned earlier for re­actor start-up and shutdown, and is relevant also in this section when considering, for example, power changes to match changes in electrical generation, power changes for refuelling (AGR), boilers being brought into or taken out of service, and gas flow faults such as blower/circulator failure.

Clearly if the gas flow changes but neutron power does not change, then temperatures must change, and normally it is desirable to avoid large temperature changes; large upward swings might trip the reactor, downward swings might shut down a magnox reactor on ’temperature poisoning’ or cause wetting of the superheater region of the boiler tubes in an AGR. Therefore the neutron power is changed to match the gas flow and thereby keep temperature changes man­ageably small.