冷冻食品（英文第二版）：34990wp_05

5.1 Introduction The practice of measuring and keeping records of temperatures is not new to the food industry, and has been undertaken by certain sectors, e.g. canning, for many years. However, its widespread application in the refrigerated food sector, other than fitting temperature measurement equipment to chill stores, is relatively recent. The prime factor focusing attention on temperature monitoring was the concern about food poisoning and the introduction of new legislation covering temperature control of chilled foods, where temperature abuse and likely growth of pathogens could be a problem. However, national changes were overtaken by developments in the European Community where a harmonised hygiene directive was developed and agreed. This coupled with vertical hygiene directives on animal based foods laid more emphasis on risk management. Thus the practice and use of temperature monitoring has matured quite rapidly over the last ten years and become more integrated into quality and safety management systems. 5.1.1 Changes in legislation Both the Food Hygiene (Amendment) Regulations 1990 1 and the Food Safety Act 1990 2 produced a significant change across the chill chain. The Food Hygiene (Amendment) Regulations 1990 1 introduced temperature controls for certain types of chilled foods which applied at all stages of the ‘chill chain’. Further minor amendments were made in 1991. 3 Up to this date very few end users of refrigerated systems practised regular temperature monitoring, however, 5 Temperature monitoring and measurement M. L. Woolfe, Food Standards Agency, London The views expressed in this Chapter are those of the author and should not be regarded as a statement of official Government policy. when they began to apply it they realised the concurrent benefits of process and quality control it brought. The Food Safety Act 1990 gave Ministers additional powers to legislate in many new areas. One significant change in the 1990 Act is found in Section 21. This describes the conditions under which a defence to charges brought under the Act can be offered. The ‘warranty’ defence under the 1984 Act has been substituted by a ‘due diligence defence’. In order to show ‘due diligence’, companies have to demonstrate that they took all ‘reasonable precautions’ and exercised ‘all due diligence’ in their operations. Many companies have moved to better systems of control and inspection on the basis of parallel case law of ‘due diligence’ in other legislation. A harmonised horizontal Hygiene Directive 93/43/EEC 4 was agreed and implemented in the UK in 1995 (The Food Safety (General Food Hygiene) Regulations). 5 These laid emphasis on a risk or HACCP (Regulation 4(3)) approach to hygiene rather than giving prescriptive or detailed rules about hygienic requirements and practices. There is a general requirement that temporary premises and equipment for transport must be capable at maintaining the food at appropriate temperatures and where necessary their design must allow those temperatures to be monitored. The requirements of the Directive relating to temperature control are enacted in the Food Safety (Temperature Control) Regulations 1995. 6 In addition, the Government was able to take advantage of the computer model developed over five years (MAFF Micromodel) to look at predictions of microbiological growth under the different temperature conditions to ensure safe food, and as a result the earlier temperature control regulations were simplified. Controls already existed in European Community legislation in the trade of animal-based foods, e.g. meat, meat products, poultrymeat, etc. In order to implement the Single Market after January 1993 some ten vertical directives were agreed dealing with the hygienic production of animal products from fresh meat to bivalve molluscs. Some of these were new and others re-negotiated from intra- community trade directives. All of these Directives have some temperature control requirements. Work is in progress to consolidate all the vertical hygiene directives and the horizontal Hygiene Directive into one simplified directive. The only mandatory requirement for temperature monitoring and keeping records is based on a European Community measure which requires monitoring equipment to be fitted to cold stores and vehicles which store or transport quick-frozen foods (Commission Directive 92/1/EC). 7 This same requirement has also been adopted in the UN/ECE agreement which facilitates cross-frontier traffic in perishable foods (ATP 8 ) in order to harmonise EC requirements for third-country vehicles. 5.1.2 Risk and quality management systems Once companies began to investigate and implement temperature monitoring systems, it soon became evident that there are derived benefits to offset the incurred capital costs and effort. Better control of temperature underpins both 100 Chilled foods the safety and quality of the food product, and can bring economic benefits of energy efficiency. The changes in legislation also necessitated the setting up of appropriate systems to ensure the safety of foods. All food businesses have the responsibility of identifying critical steps in their own processes. The approach adopted nationally and internationally is HACCP (Hazard Analysis and Critical Control Point). This identifies the risks and the critical control points in the process to control the risks. The important issue is that a HACCP plan is unique to a particular product and process, and it should be continually assessed. Help in implementing HACCP has been given by provision in the Hygiene Directive which encourages sectorial guidelines, and to date six guides 9 have been published. Temperature monitoring may or may not be part of the HACCP plan depending on the product and process involved. There is no specific requirement to keep records of temperature checks, but records may be helpful to show that legal requirements have been met. More importantly temperature monitoring is put into context with all the other control points, and integrated into the overall HACCP system. It is clear that HACCP is rarely implemented in isolation but combined with quality systems to ensure that a production unit manufactures food that is safe and is of consistent quality for the consumer. There are many systems of quality assurance, and the most widely used ones are based either on ISO 9000 or on TQM (total quality management). ISO 9000 10 series has two main standards (ISO 9001 and 9002) and various guidelines, and companies achieve accreditation when they have implemented the standards. TQM is more a cultural approach involving all members of an organisation in achieving consistent quality and consumer satisfaction and also has the concept of continual improvement. 5.1.3 Improvement in technology The ability to produce microelectronics relatively cheaply has enabled the manufacture of relatively small devices that store large amounts of data. These are now routinely integrated with computerised management systems. The last few years have seen an enormous advance in computer and communications technology. Now satellite tracking systems can follow a vehicle’s position and give total information about the refrigeration and engine systems to its depot. Retail display cases can also have integrated temperature and humidity control to ensure full shelf-life of non-pre-packed foods. Thus where temperature measurement is part of the integrated safety and quality, new technology lends itself to the storage and processing of the data. 5.2 Importance of temperature monitoring Temperature control requirements for England and Wales apply to food which is likely to support the growth of micro-organisms or the formation of toxins. Temperature monitoring and measurement 101 These foods need to be kept at 8oC or below. However, this requirement has to be implemented in combination with the others laid down in the general hygiene regulations (The Food Safety (General Food Hygiene) Regulations 1995 5 ). Obviously, if pathogenic organisms can be prevented from entering the food then temperature control is necessary only for extending the shelf-life of the product. However, this is rarely the case and the approach taken by HACCP is to identify at each stage of the preparation of a food where the hazards occur and how they can be controlled. Reducing the temperature does not kill micro- organisms, but it retards their growth. Hence keeping raw materials, intermediate and finished products at chill temperatures will play its part in ensuring the food is safe. The other important areas are proper hygiene training of operatives, prevention of physical contaminants, suitable fittings and equipment, and good cleaning regimes and pest control. Refrigeration equipment is built to function for long periods without attention, however there are many events apart from breakdown which can affect temperature control. The defrost cycles need attention to ensure they are at the correct frequency, and loading of food into refrigerated systems is often crucial to its operation and proper air flow. Air temperature monitoring can indicate whether refrigerated equipment is functioning correctly and is being operated correctly, even though it may be more difficult to extrapolate food temperatures. In some circumstances air temperature monitoring is not possible and product temperature or product simulant temperature is required. 5.3 Principles of temperature monitoring 5.3.1 Choice of system There are an enormous number of different temperature monitoring systems available commercially, from a simple thermometer to a fully computerised system linked to a local refrigeration system or even central control system. The choice of system will depend on exactly the amount of detail the operator requires and the cost at which this information is provided. If the monitoring system is to provide detailed information on the operation of a system linked with other reactive management systems, then obviously a more elaborate and complex system is required. This may include a large number of sensors to enable a very complete picture of the temperature distribution within a refrigerated system to be obtained. It may also include other information such as defrost cycles, compressor and expansion valve pressures, door openings, and energy consumption, and may be linked to an alarm system (and even telephone) stock keeping and batch codes of product. On the other hand, if monitoring is being carried out only to ensure that food is being kept within certain temperatures as a critical control point, then the amount of information which is collected may be reduced. Very little formal advice has been given in the previous literature on temperature monitoring (IIR, 11 SRCRA, 12 BRA, 13 RFIC 14 ). Guidelines 102 Chilled foods published by the IFST 15 give details about air temperature monitoring, and these were further amplified in the Department of Health’s Guidelines. 16 These Guidelines have been superseded by Industry Codes of Practice. 9 Practical advice on temperature monitoring has been included as an advisory Annex in some of the Codes (Retail and Catering Codes in particular), but these do not form part of the Codes. 5.3.2 Which temperature to monitor? When designing a monitoring system, there are certain considerations in the choice of temperatures to be measured in the refrigerated system. These are: ? The choice of whether to monitor air temperatures, product temperatures or simulated product temperatures will depend on the individual system and the way it operates. ? The sensors should preferably be fixed in a position where they will not be damaged during commercial activity. If manual readings are used, these should be taken from accessible positions. ? The temperatures chosen should be representative of the refrigerated system and give a picture of its functioning, and therefore be linked indirectly with the product temperature. 5.3.3 Air temperature monitoring In terms of regulatory compliance and as part of HACCP, the temperature of the food should be monitored. However, the storage or holding times of chilled food are relatively short, making product temperature monitoring difficult without disruption to normal commercial activity and requiring the intervention of trained operators. It is easier to fix sensors separate from food loads, which are connected to read-out systems, where temperatures can be recorded auto- matically or manually. Most refrigerated systems function by circulating cold air over the system’s evaporator, and then passing this cold air over the food load to remove heat from the food. Movement of air is by mechanical fans or in some cases gravity, which relies on density being greater for cold air than for warm air. In the case of mechanical circulation, the air returns to the evaporator after passing over the food, making the returning air the same temperature or warmer than the food it is cooling. Localised heating effects from lighting or other effects may give rise to ‘hot spots’ or uneven temperature distribution, and make a small part of the food load warmer than the return air. In general, the relationship between air temperature and product temperature is best established by examining the difference between the cold air leaving the evaporator and warmer air returning to the evaporator. This gives a measure of the performance of the refrigerated system and its effectiveness in keeping the food cold (BRA 13 ). This differential is also used as the basis of air temperature monitoring. However, in order to Temperature monitoring and measurement 103 relate the air temperatures to product temperature, it is necessary to carry out a load test. The load test involves examining the differential of air temperatures and comparing it to product temperature over a sufficient period of time to ensure the system is working under normal conditions. With closed systems such as chill stores and vehicles, where the only perturbation derives from defrost cycles, door opening and changing loads, deter- mination of the relationship between air and product temperature is simpler. The warmest locations in the system have to be determined, and product temperatures followed over a period of time in order to relate them to air temperatures. With open systems such as display cabinets, their operation is more sensitive to environmental conditions and location. Room temperature and humidity variations, perturbation of the air curtain by draughts or customer movement can change the temperature distribution. Under these circumstances, load testing can be more difficult. Cabinet manufacturers perform a load test to check cabinet’s performance (BS EN 441-5: 1996 17 ), using a set loading of standardised blocks of a gel (Tylose) (BS EN 441-4: 1995 18 ) under controlled environmental conditions of temperature with a constant air flow across the front of the cabinet. Whether the manufacturer’s load test will deviate from that in situ will depend on how close the conditions and load follow the actual working operation of the cabinet. The siting and environmental effects of draughts and lighting should be checked with a range of foods on display. 5.3.4 Alternatives to air temperature monitoring There are some circumstances where air temperature monitoring is not appropriate or needs modification. In closed cabinet systems, such as chill storage cabinets using gravity cooling from an ice-box or backplate, the air temperature inside requires significant time to recover after door openings. 19 Thus, periodic readings of air temperatures would have little meaning and bear no relationship to the temperatures of the food being stored. In this case it would be better to monitor either a food sample or a ‘simulated’ food sample. The thermal mass of the sample would make it less sensitive to rapid air temperature changes. Also it is possible to match the ‘food simulant’ to have a similar cooling factor or similar thermal diffusivity to the food being monitored. 20 The use of such monitoring would be essential for example where cooling is by conduction such as cold plate (dole plate) serving units in catering, or where air flows are low velocity (gravity-fed serve-over cabinets). Even where the system is cooled by forced air but the variations in air temperatures are large, e.g. small delivery vehicles and cabinet refrigerators, air temperature monitoring is still difficult to interpret. By increasing the response time or ‘damping’ the sensor or measuring system, the trends in air temperature can be followed, whilst removing the short-term variations. ‘Damping’ can be achieved physically by increasing the thermal mass of the sensor or electronically by alteration of the read-out circuitry. 104 Chilled foods 5.4 Temperature monitoring in practice 5.4.1 Chill storage Walk-in chill stores Walk-in stores consist of an insulated store chamber cooled by one or more fan- assisted air cooling units, depending on their size. The position of cooling units around the chamber varies, but is usually at ceiling level (Fig. 5.1). Air circulation should be designed to give proper distribution throughout the chamber, and to eliminate any ‘hot spots’ or stratification of air layers. In nearly all cases, air temperature recovery after door openings or defrost is rapid, permitting air temperature to be used as the most convenient means of monitoring. Retention of cold air can be further improved with the use of strip plastic curtains, or an air curtain above the door, minimising the ingress of warm air on door openings. The number of sensors to be used to monitor air temperatures in a chill store will depend on its size and the number of cooling units. Table 5.1 gives an indication of the minimum number of sensors related to volume of the store, with stores less than 500 m 3 being able to monitor air temperatures using one sensor. The positioning of the sensor is such that it gives an indication of the warmest air temperature and hence the warmest food in the store. This warmest location depends on the design of the store, especially the position of the air cooling unit in the store. Figure 5.2 gives an example of air temperatures during the 24-hour operation of a large chill store. The graph shows temperature variations during peak activities of movement of chilled foods in the afternoon and evening compared Fig. 5.1 Air circulation in a chill store. Temperature monitoring and measurement 105 to quieter loading activity in the morning. Differences between wall sensors and air return temperatures are very small in this case, and can be affected by their positioning in the store. For chill stores less than 500m 3 , the single sensor could be placed in the air return of the cooling unit. In a closed system such as a store with adequate air distribution, the temperature reading of the air return approximates to the mean temperature of the food load. If there is not good air distribution, then it may be better to put the one sensor in a position more representative of the warmest air temperature. This may be located at the following positions: ? the maximum height of the food load, furthest away from the cooling unit ? at approximately two-thirds the height of the chamber, away from the door and the direct path of the cooling unit ? two metres above floor level, directly opposite the cooler unit. If the cooling unit is placed above the door, the negative pressure produced by the fan can increase the amount of air drawn into the chamber during door openings. Thus, air return temperature monitoring is not often appropriate in this case. For larger stores, different sensors can be used to indicate the temperatures in different parts of the store. In addition, placing extra sensors in the air outlet and air intakes of one or more of the cooling units gives further information on the performance of the refrigeration system. Cabinet refrigerators Cabinet refrigerators are free-standing, small-sized units with single or double doors. They can be cooled by fan-assisted cold air or by gravity-circulated air from an integral icebox or backplate (Figs. 5.3, (a), (b) and (c)). As indicated earlier, air temperature monitoring is not as appropriate to these types of refrigerated systems as it is to walk-in chill stores. Fan-assisted refrigerators will recover relatively quickly after door openings, but a large number of door openings, especially at most active periods of use, will make any temperature readings difficult to interpret. Air temperature monitoring can be more meaningful if a ‘damped’ sensor, with a response time of around 15 minutes, in the air return position is used (Fig. 5.3(a)). Damping can be achieved by using a metal or plastic sheath over the sensor or suspending the sensor in water, oil or glycerol. Figure 5.4 shows the effect of ‘damping’ Table 5.1 Number of sensors recommended in chill stores Chambers volume above (m 3 ) Number of sensors 500 2 5 000 3 20 000 4 50 000 5 85 000 6 106 Chilled foods when the sensor is set at the centre of a plastic tub of water, and readings are compared to air temperatures after door openings. Since cabinets cooled by a backplate or icebox have weak air circulation and long recovery times after door openings, it is more appropriate to monitor their temperatures using food temperatures or, even better, a simulated food Fig. 5.2 Air temperature monitoring record of large chill store (40 000 m 3 ). Temperature monitoring and measurement 107 temperature. As foods are microbiologically unstable, food temperature monitoring would require using different foods each day, and might lead to wastage. Permanent positioning of a sensor requires a stable food simulant. It is important when choosing a food simulant that it behaves similarly to the food being monitored, and it is robust to different working conditions. It is recommended to determine the cooling factor of the specific package or piece of food and match this with a particular food simulant, or match the thermal diffusivity of the food to that of the simulant. 20 Values for cooling factors of different foods, package sizes are also published, 20 as well as thermal diffusivities for a range of plastic materials. Regular checks should be made with a food simulant system to ensure that the sensor embedded in it is accurate and functioning properly, and that the simulant is performing as it should. 5.4.2 Chilled transport Distribution of chilled foods is carried out in many different types of vehicle ranging from large 40-foot heavy goods vehicles with independent cooling units, to light goods vehicles relying on insulated containers to maintain temperature of pre-chilled foods. Pre-chilling to the correct temperature is essential given that most refrigeration units are designed to maintain temperature not cool the load down. A: Air off. B: Air return (air-on). C: Load limit or warmest point. Fig. 5.3 Cabinet refrigerators. (a) Forced air refrigerator. (b) Icebox refrigerator. (c) Backplate refrigerator. 108 Chilled foods Temperature controlled vehicles An independent refrigeration unit usually powered by diesel, often with an auxiliary electric motor, is used to circulate cold air around the vehicle compartment from an evaporator unit at the front of the vehicle. A trend in many multiple distribution depots is to use vehicles with movable bulkheads so that one vehicle can carry frozen and chilled foods at different temperatures in the same load. Each compartment will have its own evaporator, which can control temperatures independently. The cold air is distributed in different ways within the different vehicles, but the majority have cold air leaving from the top of the cooling unit near the roof, and returning via the base to the front of the vehicle and the return air intake (Fig. 5.5). Correct loading and spacing of the load within the vehicle is crucial to ensure adequate cold air distribution within the compartment. If the load is not spaced correctly, circulation can be restricted and ‘hot spots’ can occur. The maximum length and width of vehicles is set by regulation, and hence the free space available to loads within an insulated chamber, place further restrictions on achieving correct loading. Some vehicles are cooled by direct evaporation of liquid nitrogen from a reservoir tank on the vehicle. These vehicles have the advantage of being much quieter than mechanically refrigerated vehicles, and temperature control can be better. However, an adequate supply of liquid nitrogen is required for the journey, which can limit their range and number of stops. Fig. 5.4 Effect of ‘Damping’ air sensor. Temperature monitoring and measurement 109 Temperature read-out and single-channel chart records, which have been used for many years on refrigerated vehicles, placed the sensor so as to measure the air return temperature. This returning air should indicate the mean temperature in the load, provided that there is good distribution to all parts of the load. Short circuiting of air may result in colder return air temperatures. Long vehicles, especially those without air duct distribution of cold air in the ceiling of the compartment, are advised to fit a second sensor placed nearer the rear of the vehicle (Fig. 5.5). The addition of a second sensor is not sufficient to give a full and accurate picture of temperature distribution within the chamber, but it will measure the cold air leaving the evaporator, and may give a better picture of cold air circulation inside the compartment. This second sensor will serve as a check on the functioning of the measuring system, and makes tampering more difficult. It should demonstrate that the evaporator and fan unit are functioning properly and that cold air is reaching the back of the vehicle. It will give a temperature baseline with which to measure the return air, and indicate more easily when the cooling unit has been switched off, or a load added which has been insufficiently cooled. Prevention of freezing of part of the load can also be more easily avoided. Comparison of the normal differential temperatures between the rear sensor and the return air sensor may also indicate poor air distribution within the compartment. The frequency of recording for electronic loggers will depend on the length of the journey. A maximum interval of 15 minutes is recommended for journeys of up to 8 hours. Longer journeys may use longer intervals between recordings. Other information such as defrost cycles, door openings and load identification may be required. It is important that a driver be aware of any problem occurring with the temperature of the load. The temperature read-out is often visible to the driver in the vehicle’s wing mirror, and in some cases the read-out is presented as a mirror image. Obviously, the driver should have complete concentration on the road, and it is better that an alarm system be fitted which warns the driver when something is wrong. Fig. 5.5 Air temperature monitoring of temperature controlled vehicle. 110 Chilled foods Figure 5.6 (a) shows an example of temperature monitoring in a vehicle fitted with two sensors, including the effect of door openings. Figure 5.6 (b) illustrates the care by which air temperature records must be interpreted. The system is operating normally until the chamber is loaded. From this point, the air return sensor gives the acceptable reading apart from slightly longer cycles. Whereas the compartment sensor at the rear of the vehicle shows a temperature rise indicating that the flow of cold air has been restricted by the load. This causes short-circuiting of cold air from the evaporator and hence the longer cycling period activated by the thermostat. As soon as the driver rearranges the load to restart the air flow to the rear of the vehicle, the temperature drops. This problem would not have been obvious had there been only one sensor on the air return. The monitoring of vehicles with movable bulkheads would require more sensors to enable the temperature to be recorded in each separate compartment. This may be achieved in several ways. The easiest would be to monitor the air intake of each cooling unit. Alternatively, more sensors could be fixed to the roof of the chamber to enable compartment temperatures to be monitored, whatever the position of the bulkhead, and in addition to the air return measure- ments. The use of small temperature loggers, whose position can be changed to suit the bulkhead arrangement, may provide another solution. For vehicles cooled by liquid nitrogen, the sensors have to be positioned in order to account for any temperature gradients occurring in the chamber. Forced circulation should eliminate gradients. If fans are not used then sensors should be placed above and below the load. Fig. 5.6 (a). Normal air temperature record and (b). Poorly loaded vehicle air temperature record of chilled foods vehicle (By permission of Cold Chain Instruments.) Temperature monitoring and measurement 111 Small delivery vehicles Many light goods vehicles delivering chilled foods are fitted with refrigerated units driven from the vehicle engine or transmission. This means that cooling is not possible whilst the vehicle is stationary. For vehicles less than 3 m 3 , advances in refrigeration have allowed the retro-fitting of efficient refrigeration units powered by the vehicle’s battery. The main problem in maintaining good temperature control arises from the number of door openings and the amount of time doors may stay open whilst orders are prepared and delivered. Typical ‘High Street’ delivery patterns can result in doors being open for 40% of the working day. This can make temperature control very difficult and also the use of air temperature monitoring inappropriate. Employment of plastic strip curtains above the door can help to reduce warm air entry whilst doors are open. However, information can be obtained if the air temperature sensors are ‘damped’ by suspending them in a small bottle of liquid such as oil or glycerol. The large fluctuations are removed from the temperature records and the trends in overall chamber temperature followed. An example of this technique of monitoring is shown in Fig. 5.7. Vehicles using eutectic plates or insulated boxes to carry foods normally use a food simulant or actual food to monitor temperature during a journey. The Fig. 5.7 Air temperature record of a small delivery vehicle. 112 Chilled foods positioning would have to be as representative of the load as possible. The temperatures could be read manually, but could also be connected to a chart recorder or logging system. 5.4.3 Display cabinets The majority of chilled foods are displayed in open cabinets. Some sectors use display cabinets with closed doors; for the purposes of monitoring, these can be considered as storage cabinets (see section above on ‘Cabinet refrigerators’). The open cabinets can be divided into two main groups, multi-deck open cabinets and serve-over cabinets. Multi-deck cabinets A fan draws air from the front grille of the cabinet, where it is cooled by passing through an evaporator. The cool air emerges from the back of the shelves to cool the food, and from the top grille to form an air curtain in front of the shelves (Fig. 5.8(a)). Advances have been achieved in the design of cabinets, which include the reduction of heat gain from internal lighting, and the stabilising of the air curtain by improved design or addition of a second curtain. Thus the ease of monitoring the temperature in multi-deck cabinets will be determined by their design and operation. In principle, the differential between the air returning from the shelves and the air emerging on the shelves is an indicator of the cabinet’s performance. Positioning of sensors or manual reading of temperature is taken from the top air- curtain grille (‘air-off’) and the lower air return grille (‘air return’) (Fig. 5.8(a)). If the normal pattern of variation in air temperatures can be linked to variations in product temperature on the shelves, then air temperature monitoring can be used on a routine basis. If other factors interfere with the air temperature relationship, such as excessive radiant heat absorption, or if a relationship cannot be established between product and air temperatures, then a product or food simulant may have to be used. Figure 5.9 shows two different air temperature patterns. The first (Fig. 5.9(a)) shows regular cyclical changes in air temperature, whilst Fig. 5.9(b) indicates much steadier conditions except during a defrost cycle. In both cases, establishment of the range of air temperature with the warmest product temperatures would allow effective air temperature monitoring. 5.4.4 Serve-over display cabinets There is a wide range of cabinets in this group for displaying meat, fish, delicatessen products, pa?tisserie, cheeses and ready-to-eat products in catering establishments. In many cases the food is cooled by cold air from a refrigerated unit, but in some, especially in catering, the food is cooled by contact with a cold plate or well (dole plate) or crushed ice. The effect of radiant heat from lighting or sunlight can be more pronounced with serve-over cabinets and affect food temperature significantly. Temperature monitoring and measurement 113 Figure 5.8(b) shows a typical serve-over cabinet for retail sale of delicatessen products using fan-assisted cold air. Air emerging from the back grille cascades over the food and returns via a front grille. In the case of gravity-fed cabinets, where air enters a back grille and emerges at the base shelf, there is no air return grille. Air velocities are low in serve-over cabinets to reduce dehydration of the displayed products; this also makes air temperature measurement more difficult. Positions of sensors or manual measurements for air temperatures are also shown in Fig. 5.8(b). The relationship of air to product temperatures is necessary before air temperatures can be used routinely. Fig. 5.8 Air temperature monitoring in retail display cabinets. (a) Multi-deck cabinet. (b) Serve-over cabinet. 114 Chilled foods In many cases it will be easier to monitor the temperature of the cabinet using food temperatures or a food simulant. The temperature near the front of the cabinet will usually be indicative of the warmest locations and hence warmest foods within the cabinet. Air temperature monitoring is not suitable for cabinets Fig. 5.9 Temperature monitoring records of two different display cabinets. (By permission of the University of Bristol.) Temperature monitoring and measurement 115 cooled by conduction (dole plate or crushed ice). In this case, direct measurement of food temperatures is appropriate, but should be carried out, as with all determinations of this type, with a clean, well-disinfected probe. 5.5 Equipment for temperature monitoring 5.5.1 European Standard for temperature recorders and thermometers In view of the multiplicity of national specifications and testing procedures for temperature recorders and thermometers for the transport of quick-frozen foods in Member States of the European Union under Directive 92/1/EC, 7 and the growing importance of temperature control in chilled food and frozen food sector, the European Centre for Standards (CEN) has agreed a standard for temperature recorders. One Standard, BS EN 12830: 1999, 21 and one draft Standard prEN 13485, 22 cover temperature recorders for the transport, storage and distribution of chilled, frozen, quick/deep-frozen food and ice cream and thermometers in the same field of application respectively. There is a third draft Standard, prEN13486 23 which lays down the procedure for periodic verification of the temperature recorders and thermometers in the preceding standards. Temperature recorders The CEN Standard 21 lays down requirements for electrical safety, resistance to mechanical vibrations, and operational performance in climatic conditions. It also sets a minimum specification for accuracy, response time, recording interval and the maximum relative timing error. Table 5.2 gives the climatic environments under which a temperature recorder and a thermometer measuring air temperature must operate even when stored or operated for short periods under those conditions. Obviously these vary as to whether the device is operating inside the refrigerated chamber or at the vagary of outside weather conditions compared to a device operating inside a building or vehicle. The CEN Standard 21 also lays down the test conditions under which the specification for temperature recorders are determined. Thermometers The draft Standard prEN13485 22 defines a specification for thermometers for measuring air temperature in transport, storage and distribution and product temperature of chilled and frozen food. Table 5.2 gives the environmental conditions under which thermometers measuring air temperature have to operate under different uses and Table 5.3 their response times. Table 5.4 gives the environmental conditions under which portable thermometers for air tempera- ture and product measurement must operate. For product thermometers there is also a limit of 0.3oC change in accuracy when operating across the full range of ambient temperatures (C020oC to +30oC). The accuracy classes for thermometers measuring air and product temperature are given in Table 5.5. The draft 116 Chilled foods Table 5.2 The climatic environments under which a temperature recorder and a thermometer measuring air temperature must operate Recorder or thermometer for storage and distribution, located outside refrigeration case in heated or air conditioned premises and with external sensor Recorder or thermometer for transport located in or outside vehicle with external sensor Recorder or thermometer for storage and distribution, located inside refrigeration case and with external or internal sensor Recorder or thermometer for transport located inside refrigeration case with internal or external sensor Thermometer, recorder and displaying device rated operating conditions* +5oC+40oC C030oC+65oC C030oC+30oC C030oC+30oC Thermometer, recorder and displaying device limiting conditions C121 0oC + 50C C030oC+70oC C040oC+50oC C040oC+70oC Thermometer, recorder and displaying device and sensor storage or transport conditions C122 C020oC+60oC C040oC+85oC40oC+60oC C040oC+85oC * Conditions which device operates to specification. C121 Conditions which device can withstand whilst in operation so that it will subsequently operate according to specification when under its rated conditions. C122 Conditions which device can withstand whilst not operational so that it will subsequently operate according to specification when under its rated conditions. Table 5.3 Response times* for the sensors in temperature recorders Type of device Transport Storage All uses External sensor 10 min. max. 20 min. max. Internal sensor ––60 min. max. Fixed thermometers 10 min. max. 20 min. max. Portable thermometers 3 min. max. Product temperature thermometer ––3 min. max. * Response time is the time needed for the measured or recorded value to reach 90% of the actual change of applied temperature under test conditions Temperature monitoring and measurement 117 Standard 22 also lays down the test procedures for determining temperature measuring error and response time. 5.5.2 Sensors Accuracy Whatever the system for collecting or recording temperatures, the sensor or heat-sensitive part is the common factor between them. The three principal types of sensors in commercial use are thermocouples, platinum resistance and semiconductor (thermistor). The choice of which type of sensor will depend on the requirements for accuracy, speed of response, range of temperatures, robustness and cost. Until recently, the majority of general-purpose thermometers and measuring systems used a thermocouple in the heat-sensitive part of the system. This is a pair of dissimilar metals joined together at one end, usually by a soldered joint. The circuit is completed by a second junction held at a known temperature (often referred to as a ‘cold’ junction). In food applications, where temperatures are relatively close to ambient, two types of thermocouple predominate: Type K thermocouples, which use wires of Chromel (a nickel–chromium alloy) and Alumel (a nickel–aluminum alloy); and Type T thermocouples, which use wires of copper and Constantan (a copper–nickel alloy). The advantages of the thermocouples are their low cost, facility to be hand-prepared from reels of wire, and a very wide range of temperature measurement (C0184oC to 1600oC). Table 5.6 shows the permissible sensor accuracies for the three types of sensors, which in the cases of thermocouples and the platinum resistance sensor conform to a standard requirement. The difference in instrumental error arises Table 5.4 Climatic environment under which portable thermometers and thermometers for product temperature must operate Thermometers for product temperature Rated operating conditions C020oC+30oC* Limiting operating conditions C030oC+50oC Storage operating conditions C030oC+70oC * For measurements made in this ambient temperature range, the measuring accuracy shall not change more than 0.3oC. Table 5.5 Accuracy classes for thermometers measuring air or product temperatures Air temperature Product temperature Class 1 2 0.5 1 Maximum possible errors C61oC C62oC C60.5oC C61oC Resolution C200.5oC C201oC C200.1oC C200.5oC 118 Chilled foods from the fact that the electronic circuitry has to be compensated for changes in the reference or ‘cold junction’ (normally ambient temperature). This is measured by a built-in semi-conductor sensor, which automatically compensates for changes in ambient temperature. Errors using thermocouples increase when the ambient temperature varies widely, i.e. moving from a cold to a hot environment. Other errors with thermocouples can be produced by induced voltages from motors or transmitters, moisture, and thermal gradients in other junctions. A move to greater accuracy for measuring and monitoring restricts the use of thermocouple sensors to Type T only, which will normally meet the basic specification for air temperature monitoring (CEN Standard 21 ). Thermistor sensors change resistance with temperature, but normally can be used to measure only a narrower range of temperatures than the thermocouple (C040oC to 140oC). Their use for measuring food temperatures has increased since the introduction of a basic specification for measuring systems for food temperatures to give an accuracy of C61oC, which is reinforced by the draft CEN Standard for thermometers. 22 They are rugged, provide good accuracy and repeatability and are not unduly affected by changes in ambient temperatures. Platinum resistance thermometers also have a system accuracy which meets the draft CEN Standard. 22 They may be used over a wider range of temperatures (C0270oC to 850oC). Normally their response time (Table 5.7) is slower, unless they are specially constructed for a fast response. Corrections have to be made for resistance of leads, and the self-heating effect. Their higher cost has restricted their use to applications where high accuracy is required in fixed process control applications. Calibration and periodic verification During manufacture, each sensor and instrument is checked to ensure that it meets specification and achieves an accuracy within tolerances set by each manufacturer and in accordance with BS EN 12830:1999 21 and prEN 13485:1999. 22 In many applications of monitoring, different sensors are plugged into an instrumentation system, and they are normally regarded as interchange- able. However, in the case where more precise readings are required, an individual calibration is undertaken of the sensor and instrument together Table 5.6 Sensor and system accuracies Type K Type T Pt resistance Thermistor Sensor accuracy (oC) C61.5* C60.5* C60.2 C121 C60.1 Instrument accuracy C122 (oC) C60.3 C60.3 C60.2 C60.2 System accuracy (oC) C61.8 C60.8 C60.4 C60.3 * BS 4937: Class A 24 C121 BS 1904: Class A 25 C122 Includes cold junction compensation accuracy. 26 Temperature monitoring and measurement 119 (system). This measures the system’s reading against a range of applied temperatures. The applied temperatures have to be traceable to a national standard (e.g. National Physical Laboratory). The resulting table or graph of the calibration certificate enables the system reading to be corrected to a true reading (within the tolerances of the calibration). Once a temperature monitoring system is installed, it is essential that periodic checks are carried out to ensure that the equipment is functioning correctly and meets the same specification as when it was purchased, and as described in prEN13486:1999. 23 The frequency of checks depends on the use of the equipment and will consist of routine checks on the functioning of the equipment and those carried out by the manufacturer (or a suitably qualified laboratory). The maximum period recommended is one year for a manufac- turer’s check or after a long period of non-use or operating incident. The equipment is normally checked against another thermometer, which has been calibrated against a standard. It is normal to check the accuracy, and functioning of the clock or verify the recording duration. Sensor housing and probes In monitoring applications, the sensor element has to be protected from damage or breakage. This can range from coating with an epoxy resin to embedding in a stainless-steel sheath. If fast response is required, the thermal mass has to be as low as possible. It is also important that sensors, which are mounted inside chambers or vehicles to measure air temperatures, are also protected from damage during commercial activity of loading or unloading food, but not in such a way as to restrict air flow. Monitoring and measuring food temperatures often requires sensors mounted in hand-held probes. The design of the probe depends on its application. The most common probe is for insertion into foods, and therefore has a sharpened tip (see Fig. 5.10(a)). If a non-destructive temperature measurement is required then a probe is required which can be inserted between food packs or cases. Good contact between the packaging and the probe, together with an adequate period to allow readings to settle, are essential to minimise errors in this type of measurement. Examples of probes used to measure between packs and cases are shown in Figs 5.10((b) and (c)). Table 5.7 Typical response times (seconds) in air and water 26 Still air Forced air Water* Exposed thermocouple 20 5 – Shrouded thermocouple 150 40 6 Exposed thermistor 45 20 – Shrouded thermistor 260 50 12 Shrouded platinum 365 65 15 * Mounted in ‘chisel’ probe in water, time for 20oC change to 99% level. 120 Chilled foods 5.5.3 Read-out and recording systems Single readout systems Instrumentation has progressed from the original single read-out thermometer, the mercury- or alcohol-in-glass thermometer. The development of dial and stick Fig. 5.10 Hand-held temperature probes. (a) Various air and product temperature probes. (b) A probe for between-pack temperatures. (c) A probe for between-case temperatures. Temperature monitoring and measurement 121 thermometers with analogue or digital display removes the danger of breakage, but their use can be limited by their low accuracies especially those based on bi- metallic strips. Dial thermometers which were used to indicate air temperatures in display cabinets have largely been replaced by digital thermometers. Thermochromic liquid crystals change in orientation and transparency depending on their composition and temperature. When set into strips they will display the appropriate temperatures printed under them. Their accuracy is limited, but can achieve C61oC. More common is the electronic digital readout instrument, which is powered by batteries. The resolution and interval of display temperature will vary with model and type of sensor. Temperatures can be stored and even printed out, and an alarm given if the temperature goes outside a preset limit. Chart recorders Historically, a trace on a moving chart was the only method available to produce a temperature history and record. The use of chart recorders is less common and they have been overtaken by electronic instruments but some may still be found in fixed system applications such as cold or chill stores and vehicles. The charts can be circular or mounted on a roll to give a rectangular chart, and a trace obtained with ink or pressure or heat-sensitive paper. Circular chart recorders have the advantage that the temperature history is visible and abrupt changes apparent, and the chart can be easily stored for future reference. The timescale of the chart is usually over 24 hours, 7 or 31 days, but some long- distance marine recorders may operate for 6–8 weeks. The chart clock and electronics can either be battery driven for mobility or driven from the mains for fixed applications. The accuracy of the recording duration according to BS EN 12830: 1999 21 has to be 0.2% of the recording time when less than 31 days and 0.1% of the recording time when over 31 days. Accuracy of the system varies with sensor, but more modern chart recorders are better than 0.5oC in the range 0–25oC. Their limitation is often the resolution on the chart divisions and the thickness of the trace. Fixed system charts can be sophisticated instruments with the possibility of recording 30 or more different channels in different colours and print modes. Chart recorders which are fitted to vehicles (or more often to the trailers) have to be more robust in their construction. They need to withstand the rigours of the road for all types of terrain and weather conditions. Recorders giving two or more traces are available, and event markers (e.g. noting door openings) can be added. Fixed processing system for chill stores The difficulty in interpretation of a large number of different traces and the rapid developments in microelectronics and computer technology have encouraged the replacement of chart recorders by data logging systems. This allows not only the storage of large amounts of data, but also its manipulation and analysis and integration into management systems. 122 Chilled foods In chill store operations where large numbers of temperature measurements are being taken every day throughout the year, it has become increasingly the practice to install computerised systems to handle the data. There may be a digital temperature display on a control unit situated near the refrigerated system, but more often information can be retrieved on a visual display unit situated in a control room. Alarm systems can be integrated with the system when any of the parameters being monitored are outside preset limits. The alarm can be transmitted to maintenance staff inside the system’s locations or to outside premises through telecommunication networks. Vehicle temperature logging systems Several companies have developed dedicated temperature logging systems for vehicle monitoring. These have been designed to withstand the vibration and harsh conditions encountered in transport as stipulated in BS EN 12830: 1999. 21 Data is collected over the whole journey from loading to unloading, and alarm signals given if temperatures are outside preset limits. The equipment can either fit inside the vehicle cabin and is often the same size as a vehicle radio, or will be fitted outside the vehicle often next to the refrigeration control unit. In addition, distribution customers are increasingly requiring a record of the temperature history of the food they receive. Systems have been developed which give an instant print-out of temperatures up to the point of delivery, to be attached with the delivery documentation. Other features found are ability for variable logging periods and up to 12 months memory, event recorders for defrost and door openings, and multiple channels for multi-compartment monitoring. The retrieval of the information is becoming more sophisticated with downloading facilities to office PCs via manual collection units or by radio, infrared, or satellite communications. Portable data logging systems The miniaturisation of circuitry has produced some very compact and powerful data logging systems, some of which are small enough to travel with food cases or pallets and record temperatures during passage through the chill chain. The devices can also be used as fixed systems in stores and vehicles. This is useful where the position of fixed sensors would have to be changed from time to time, e.g. temporary chill stores or movable partitions in multi-compartment vehicles. The choice of system will then depend on its particular application, convenience of its use, and price. An evaluation of two such devices has been reported by Kleer et al., 27 where they were used in large-scale catering systems, and found useful in recording the critical control points of the process. Another type of data logger (‘electronic chicken’) is useful for monitoring display cabinets. The logger is placed on a shelf and then records temperatures from a food simulant contained in the logger, which has the same thermal properties as the food displayed on the shelf. An alarm light is fitted to the logger so that problems can be easily noticed and remedied. The data is down- loaded for display and analysis via an infrared remote reader. Temperature monitoring and measurement 123 The specification of commercial systems is changing as micro-electronics improve, and it is certain that miniaturisation of the loggers will continue. Most of the present systems are still too big to fit into cases without removing one of the food packs; much smaller and slimmer devices would be able to fit between food packs. Remote sensing devices – non-contact thermometers All objects emit energy at temperatures above absolute zero. This energy increases in intensity, but decreases in wavelength, with temperature. In the temperature range of interest to chilled foods, infrared radiation can be measured to determine the temperature. As temperature increases, the intensity increases and the peak energy moves to shorter wavelengths. Hence, most low- temperature, commercial, infrared thermometers filter a band (8–14 microns) out of the infrared spectrum and measure its intensity. Using such a band reduces atmospheric (water vapour, carbon dioxide) absorption and ‘distance sensitivity’ of the instrument. Very narrow bands (2.2, 5.2 and 7.9 microns) can be used to give greater accuracy at very high temperatures, but signals are very low requiring expensive high-gain amplifiers. Not all materials emit the same energy at the same temperatures. The ratio of the energy radiated by a material compared with a perfect radiator or blackbody is known as the ‘emissivity’. Emissivities vary from 0 to 1.0, with most organic substances having a value around 0.95. Different substances vary in the amount of energy they absorb, reflect and transmit. Infrared thermometers have emissivity compensators which have to be set for different values (0.1–1.0) to allow for these differences. The target size is also important. The instrument averages all the temperatures it sees in its field of vision. Unless the object fills all the field of vision, the temperature reading will be an average of the object and its surroundings. Focal distances vary with machine, from close-up to 50 metres. The further away, the more difficult it is to pin-point targets, and laser sighting is common on many models. There are two main types of remote sensing equipment. In one type, pistol- shaped instruments are pointed at a target, and the temperature is read from a digital readout at the back of the instrument. Laser targeting can be built into the pistol to give through-the-lens sighting to locate the target, and long-distance devices are often assisted by optical telescopic sighting. Accuracies claimed for this type of instrument are around C61oC. Research carried out by the University of Bristol 28 on nine commercially available infrared thermometers revealed that care must be taken in using and interpreting results from these devices. Surface temperatures may be quite different from the internal food temperature. This is a more acute problem with frozen foods where the differential between the surface temperature and interior temperature can be large especially when the food is being transferred in a higher ambient temperature than C018oC. The infrared sensor not only measures the radiation emitted from the surface as a result of its temperature, but also radiation which is reflected from the surroundings of the food, e.g. lighting. 124 Chilled foods Depending on the type of packaging, the reflected radiation can be quite considerable and hence will give an incorrect surface temperature. There was a large difference between the performance of the nine devices when used in a commercial retail cabinet in a retail outlet. Table 5.8 shows the performance of these using six different packaging materials. Infrared readings were compared with a calibrated thermocouple inserted under the surface of the pack. Out of the five instruments, two (b & g) had errors less than 1oC, and five less than 2.5oC, with two further instruments giving unsatisfactory errors. The highest errors of all of the instruments were found with a printed foil pack which gave the most reflected radiation. It is recommended that angled and brightly lit packs be avoided with an infrared thermometer and only horizontal or vertical positioned packs in a cabinet chosen, with the device perpendicular to the top surface. In order to improve the accuracy, lighting must be reduced as much as possible, and distances to take readings must be as short as possible with as consistent a product as the situation allows. If the thermometer is moved from one ambient temperature to another, e.g. room temperature to a chill store, then for best repeatability of measurements it is advisable to allow at least 30 minutes for the instrument to adjust to the new ambient temperature. The thermometer should also be checked regularly against surfaces of known temperature. It is possible to make a relatively cheap black body calibration chamber with black PVC plastic tubing and a copper block. Alternatively commercial systems are available. The other type is based on infrared video camera-type instruments. Thermal images are displayed on a video display unit, either in colour or monochrome. A temperature scale on the display gives the temperature which corresponds to the individual colour or shade. Cameras range from low-resolution types, often used for locating victims trapped in collapsed buildings, to very high-resolution, sophisticated systems which allow computerised manipulation of the data. Infrared systems have been found to be very useful for industrial control and Table 5.8 Mean error in oC with standard deviation (in parentheses) of different packaging materials Instrument Clear Glossy Plastic Printed Printed Printed Moduli MAP Cardboard bag laminate MAP vacuum average foil pack a 0.6 (0.1) 1.7 (0.1) 1.1 (0.6) 6.6 (0.6) 1.9 (0.5) 1.3 (0.1) 2.2 b C00.3 (0.0) 0.7 (0.0) 0.8 (0.6) 5.3 (0.6) 0.6 (0.1) 1.4 (0.1) 1.5 c 0.7 (0.1) 0.6 (0.0) 0.5 (0.0) 6.0 (0.1) 0.4 (0.6) 0.4 (0.0) 1.4 d C03.3 (0.3) C04.5 (0.5) C05.1 (0.4) 7.0 (0.2) C09.1 (1.0) C07.2 (0.2) 6.0 e C01.9 (0.6) C02.3 (0.1) C02.5 (0.0) 4.1 (0.0) 1.8 (0.1) C00.6 (0.1) 2.2 f 0.8 (0.1) 0.9 (0.4) 1.0 (0.5) 4.2 (3.0) 2.9 (0.3) 2.3 (0.1) 2.2 g C00.1 (0.1) C00.5 (0.6) 0.4 (0.6) 6.2 (0.5) 0.2 (0.2) 0.6 (0.1) 1.3 h 0.5 (0.4) 3.8 (0.3) 6.4 (0.7) 10.4 (0.9) 6.1 (0.8) 4.0 (1.4) 5.5 i C02.2 (0.0) C01.2 (0.0) C00.8 (0.6) 3.2 (0.0) C00.9 (0.6) C01.0 (0.6) 1.5 Mod. Av. 1.4 1.3 2.1 5.9 2.7 2.1 Temperature monitoring and measurement 125 energy efficiency, with the ability to pick out overheating components and heat losses. They are also being used increasingly in on-line applications in the food industry, either with the hand-held type or with specific models available for this. For example, sealing rollers on microwavable plastic trays can be monitored to ensure evenness of heating, and products emerging from cooking, heating or cooling tunnels can also be screened for evenness of heating or cooling. These are applications where the target is consistent and relative temperatures are more important than accurate temperatures. The readings are instantaneous and the information can be linked directly to control systems. Infrared temperature measurement will never replace electrical temperature measurement for accurate determinations for enforcement of temperature control legislation. However, there are exciting possibilities for its use in routine monitoring and temperature auditing where relative temperatures are very important, and care needs to be exercised in the interpretation of results. Hand-held devices can be used to monitor the surface temperature of cases unloaded from a vehicle on an acceptance or rejection basis or scan a display cabinet for ‘hot-spots’. 5.6 Temperature and time–temperature indicators 5.6.1 Performance of TTIs Temperature monitoring has been discussed in terms of displaying temperature readings of the surrounding air or of the food or simulated food itself. However, it is possible to use a physico-chemical mechanism and a resulting colour change to display (a) a current temperature, (b) the crossing of a threshold temperature, or (c) an integration of the temperature and the time of exposure to temperature after activation. Such devices are called temperature indicators (TIs) in the first two cases or time–temperature indicators (TTIs) in the last case. The indicators are normally integrated onto a packaging material which can be attached to the food packaging or the outside of the surrounding or bulk packaging, and can follow the food throughout the chill chain. The type of information than can be provided is one or more of the following: ? reject or accept on the basis of a colour change ? temperature abuse above a threshold temperature ? partial time–temperature history above a threshold temperature ? full time–temperature history linked to shelf life. In order that the devices can be used in commercial situations for monitoring, they should have the following features and be supplied with the following information by the manufacturer. 29 ? ease of application to food packs; ? instructions on the activation of the TTI just before application, including the temperature at which the device must be stored if it has to be kept at low temperatures (frozen) once manufactured until applied; 126 Chilled foods ? for TIs, the threshold temperature and its tolerance limits (3C2standard deviation) in oC, and a response time (inertia) in minutes; ? for TTIs the maximum and minimum temperature limits in oC over which the device will function and the time to end point with the tolerance at sufficient numbers of temperatures throughout the range stated by the manufacturer (above the critical reference temperature in the case of partial history TTIs). The number of temperatures and time to end point combinations shall not be less than five; ? the performance tolerance for the time to end point as in BS 7908: 1999 29 : Category A (up to C62.5%), Category B (up to C65%), Category C (up to C610%), Category D (up to C620%); ? for partial history TTIs the critical reference temperature, i.e. the temperature at which the physico-chemical change is activated to produce an irreversible change in oC and its tolerance; ? the storage conditions of these devices should be given so that their performance is unaffected. Also any conditions which could affect the performance apart from temperature, e.g. light should be stated; ? the devices should be tamper-proof. It is important to appreciate that TTIs are based on physical changes, or chemical and biochemical reactions. Their performance does not in general mimic microbiological changes in food but the biochemical and chemical reactions which cause deterioration in the sensory quality of foods. Normally, biochemical reactions change at a faster rate than chemical ones. However, each food will have a different combination of reactions and hence a different rate. In designing an indicator, it may be important that the activation energy of the device is similar to that of the food deteriorations as well as rate of deterioration. 30, 31 Over 100 patents have been filed on processes which could be used as a basis for indicators. These include changes with temperature based on melting-point temperature, enzyme reaction, polymerisation, electrochemical corrosion, and liquid crystals. The result of the change is usually a colour difference, which can be represented as a static change or moving band. Available temperature and time–temperature devices have been reviewed. 32, 33, 34 Many temperature and time–temperature indicators have been launched commercially over the past 10– 15 years, but very few have survived. Therefore only a few of the more successful devices will be described. 3M MonitorMark TM These TTIs consist of a paper blotter and track separated by a polyester film layer. The blotter is impregnated with chemicals of specific melting point and a blue dye. If a pre-determined threshold temperature is reached the chemicals melt, and hence the devices are partial history TTIs. The indicator is activated by Temperature monitoring and measurement 127 removing the film, the chemicals and dye diffuse along the blotter and track. There are five windows which, as they turn blue, allow the exposure to temperature to be estimated. The diffusion rate increases with temperature above the melt point. Varying tags are produced to correspond to different lengths of time at different melt temperatures (these range from C015oCtoC631oC). Lifelines Lifelines has developed several indicators all of which show a full time– temperature history. The indicator part consists of polymeric compounds that change colour as a result of accumulated temperature exposure. The colour change is based on polymerisation of acetylenic monomers which proceeds faster at higher temperatures leading to a more rapid darkening of the indicator. In the first type of label, there are two parts; a standard bar code and an indicator bar. A portable computer with an optical wand is used to read the indicator, giving a reflectance reading. Initially reflectance is high (95–100%); this falls during use, as the reaction proceeds and the colour darkens (50% reflectance). The computer correlates the change in reflectance to the time– temperature characteristics. This is linked to the information about the product in the bar code, and predictions on the quality life can be made. Developments in Lifelines technology have resulted in the manufacture of the ‘Fresh-Check’ indicator, designed for consumer use. This device consists of two circles; C213 a small inner circle which contains the polymer, and a printed dark or black outer ring. The inner ring darkens when exposed to time and temperature combinations at a rate predetermined by the durability of the food. The consumer is advised not to consume the product when the inner ring has become darker than the printed outer one (see Fig. 5.11). In order to try to link the indicator to microbiological safety as well as quality deterioration, a new development has been the incorporation of a second polymer system in the centre ring. If the indicator stays below a pre-set maximum then the polymer will change colour as above, and the change is linked to the durability. If the temperature rises above the maximum, then the second system starts to polymerise, and the centre will darken abruptly at the end of a predetermined length of time. Lifelines labels are not physically activated, and once manufactured respond to any temperatures to which they are exposed. Therefore before use, indicators must, at all times, be stored at C018oC or colder. Vitsab C213 A range of TTIs are manufactured by Visual Indicator Tab Systems in Sweden and are based on an enzymatic release of protons which changes the colour of a pH indicator from green to yellow. The rate of release is temperature related and the rate can be varied to match the shelf-life and temperature of chilled and frozen foods. The indicator can be stored at room temperature and is activated 128 Chilled foods by pressure which breaks an internal pouch allowing the components to mix. The circular indicator can be printed on flexible or semi-rigid packaging and can be incorporated into or positioned on the seal of the packaging. Activation can be post-sealing or during sealing. The TTI is also available mounted on a card which can either be placed between packs on a pallet or inside a bulk pack. 5.6.2 Practical use of time–temperature indicators (TTIs) There are some technical difficulties in the use of TTIs compared to other methods of monitoring. The fact that most are applied to the outside of a food pack means that surface temperatures are being used to change the indicators. As long as food packs are in cases, this is probably still a good guide to the food temperature with a tolerance. However, for foods on display, this may give a false indication of shelf-life, owing to radiant heat absorption, unless the effect is eliminated or compensated for. Therefore, as a means of following the integrity of the chill chain from manufacture up to the point of display, TTIs may have a practical advantage in use over certain other types of monitoring in that they give a simple and individual indication of temperature abuse. A survey of 511 consumers, carried out by the National Consumer Council, 35 indicated that almost all respondents (95%) thought that TTIs were a good idea, but only grasped their concept after some explanation, indicating that substantial publicity or an education campaign would be required. Use of TTIs would have to be in conjunction with the durability date, with clear instructions about what to do when the indicator changed colour. The relationship and possible conflict between the indication of the TTI and the durability date on the food was considered a problem. In the retail situation, nearly half those questioned would trust the TTI response if it had not changed but the product was beyond its durability date. If the TTI changed before the end of the durability date when stored at home, the majority of respondents (57%) MT70MT114MT101MT115MT104MT45MT67MT104MT101MT99MT107 MT73MT110MT100MT105MT99MT97MT116MT111MT114 MT68MT111MT32MT110MT111MT116MT32MT117MT115MT101MT32MT105MT102 MT99MT101MT110MT116MT101MT114MT32MT105MT115MT32MT100MT97MT114MT107MT101MT114 MT116MT104MT97MT110MT32MT114MT105MT110MT103 MT82 MT70MT114MT101MT115MT104MT45MT67MT104MT101MT99MT107 MT73MT110MT100MT105MT99MT97MT116MT111MT114 MT68MT111MT32MT110MT111MT116MT32MT117MT115MT101MT32MT105MT102 MT99MT101MT110MT116MT101MT114MT32MT105MT115MT32MT100MT97MT114MT107MT101MT114 MT116MT104MT97MT110MT32MT114MT105MT110MT103 MT82 MT70MT114MT101MT115MT104MT45MT67MT104MT101MT99MT107 MT73MT110MT100MT105MT99MT97MT116MT111MT114 MT68MT111MT32MT110MT111MT116MT32MT117MT115MT101MT32MT105MT102 MT99MT101MT110MT116MT101MT114MT32MT105MT115MT32MT100MT97MT114MT107MT101MT114 MT116MT104MT97MT110MT32MT114MT105MT110MT103 MT82 MT70MT114MT101MT115MT104MT45MT67MT104MT101MT99MT107 MT73MT110MT100MT105MT99MT97MT116MT111MT114 MT68MT111MT32MT110MT111MT116MT32MT117MT115MT101MT32MT105MT102 MT99MT101MT110MT116MT101MT114MT32MT105MT115MT32MT100MT97MT114MT107MT101MT114 MT116MT104MT97MT110MT32MT114MT105MT110MT103 MT82 MT70MT114MT101MT115MT104MT45MT67MT104MT101MT99MT107 MT73MT110MT100MT105MT99MT97MT116MT111MT114 MT68MT111MT32MT110MT111MT116MT32MT117MT115MT101MT32MT105MT102 MT99MT101MT110MT116MT101MT114MT32MT105MT115MT32MT100MT97MT114MT107MT101MT114 MT116MT104MT97MT110MT32MT114MT105MT110MT103 MT82 MT83MT116MT97MT114MT116 MT77MT105MT100MT45MT112MT111MT105MT110MT116 MT80MT97MT115MT116 MT111MT114 MT111MT118MT101MT114 MT116MT101MT109MT112MT101MT114MT97MT116MT117MT114MT101 MT69MT110MT100MT45MT112MT111MT105MT110MT116 MT87MT97MT121MT32MT112MT97MT115MT116 MT111MT114 MT98MT97MT100MT108MT121 MT97MT98MT117MT115MT101MT100 MT68MT97MT116MT101 MT111MT102 MT112MT97MT99MT107MT105MT110MT103 MT65MT112MT112MT114MT111MT120MT105MT109MT97MT116MT101MT115 MT116MT111MT146MT115MT101MT108MT108MT45MT98MT121MT146 MT100MT97MT116MT101 MT145MT85MT115MT101MT45MT98MT121MT146 MT111MT114 MT145MT99MT111MT110MT115MT117MT109MT101MT45MT98MT121MT146 MT100MT97MT116MT101 MT68MT111MT32MT32MT32MT45MT32MT32MT32MT78MT111MT116MT32MT32MT32MT45MT32MT32MT32MT85MT115MT101 Fig. 5.11 The Lifelines ‘Fresh-Check’ indicator – staged examples. Temperature monitoring and measurement 129 would use their own judgement in deciding whether a food was safe to eat, with at least 25% putting some of the blame on the food suppliers. However, the value of TTIs was recognised for raising confidence in retail handling, and improving hygiene practices when food is taken home and stored in refrigerators. Concerns over their technical performance (accuracy and reproducibility), were expressed, and the question of whether they could be tampered with or interfered with was also raised. These concerns are shared by the food industry, and have been addressed by the publication of a technical specification for time–temperature indicators. 29, 36 Reluctance by retailers to use an indicator on retail packs for consumer use is understandable because of the difficulties that are raised. To date, no permanent commercial use of TTIs on retail packs has been adopted in the United Kingdom. However, both French and Spanish supermarket chains have used Lifelines Fresh- Check indicators on selected items of chilled food for quite extended times, but have decided not to use them on a long-term basis. TTIs have found more extensive use in the medical field to ensure that vaccines and medicines are transported and stored correctly. In addition their use to ensure the integrity of the chill chain up to retail sale by having indicators on outer carton or pallets as a further check is being examined by the chill and frozen food industry and retailers. TTIs advantages over other types of monitoring equipment of giving easy and clear answers to whether temperature abuse has taken place makes them an attractive addition to assuring safety and quality to the consumer. 5.7 Temperature modelling and control The use of computer modelling as an aid to predict what is happening in complex systems is well established and has been applied in the refrigerated food sector. It has enabled food temperatures to be predicted if the conditions of use of the refrigerated system are known. 5.7.1 Short-distance delivery vehicles The difficulties in monitoring and maintaining food temperatures in small delivery vehicles with many stops at retail outlets have been examined by the University of Bristol, 37 and a commercial computer programme (CoolVan 38 ) has been developed to aid the design and operation of these vehicles. The programme examines the changes to the temperature of the air inside the vehicle. The ingress of heat through the insulation from the outside air and solar radiation are taken into account as is the infiltration of air through the back door when the vehicle is travelling and stationary with the door open. The thermal properties of the new insulation of the vehicle and the age of the vehicle enables the reduced heat transfer coefficient of the walls to be predicted, and each side of the vehicle can be treated separately. The infiltration of air during door openings is the one of major factors in heat gain. 130 Chilled foods Transparent plastic strip curtains have been recommended as a way of reducing ingress of air, and measurements showed that blowing air from the cooling system directly at the curtains helped to counteract warm air entering the gaps at the top of the curtain. At each stage of the programme’s development it was tested against measured data. The programme was able to predict the mean temperature of the food inside the vehicle with an accuracy better than 1oC at any time throughout the journey. However, food temperature within the vehicle actually varied by more than 5oC at one time due to the uneven temperature within the vehicle. 5.7.2 Retail display Computer modelling has been developed to examine the way retail cabinets behave and so improve their design. Computational fluid dynamics (CFD) is a tool which enables changes to be made to the computer model and see which effect produces the best results before checking this against actual measure- ments. It has been applied to study the effects of retail cabinets on supermarket environments, and in particular the cold air spillage from frozen food cabinets to the aisles. 39 The model predicted temperatures at floor level of between 5oC and 15oC, whereas measured values ranged from 13oCto22oC. The model is better at showing trends than actually using it to predict or follow actual temperatures. There is obviously an exciting future ahead where computer models will be used to improve the design of all refrigerated equipment in the food chain and improve energy efficiency whilst maintaining food temperatures. 5.8 Further reading JABLONSKI J R, TQM implementation. In: Implementing Total Quality Manage- ment: an Overview, San Diego, Ca, Pfeiffer and Co., 1991. WEBB N B and MARSDEN J L, Relationship of the HACCP system to Total Quality Management. In: HACCP in Meat, Poultry and Fish Processing, Edited by Pearson A.M. and Dutson T.R. Advances in Food Research Vol. 10, 1995 Blackie Academic and professional, Chapman and Hall. JAMES S.J., Controlling food temperature during production, distribution and retail, New Food 1999 1 (3) pp. 35–45. FRPERC Publication No. 584. MAFF Awareness Initiative: Computational Fluid Dynamics for the Food Industry, University of Bristol, 1994. 5.9 References 1. Food Hygiene (Amendment) Regulations 1990, SI 1990 No. 1431, London HMSO. 2. Food Safety Act 1990, Chapter 16, London, HMSO. Temperature monitoring and measurement 131 3. Food Hygiene (Amendment) Regulations 1991, SI 1991 No. 1343, London HMSO. 4. Council Directive 93/43/EEC on the hygiene of foodstuffs. OJ No L 175, 19.7.93 pp. 1–11. 5. The Food Safety (General Food Hygiene) Regulations 1995, SI 1995 No. 1763, London HMSO. 6. The Food Safety (Temperature Control) Regulations 1995, SI 1995 No. 2200, London HMSO. 7. Commission Directive 92/1/EEC on the monitoring of temperatures in the means of transport, warehousing and storage of quick-frozen foods for human consumption. OJ No.L 34, 11.2.92, pp. 28–9. 8. Agreement on the International Carriage of Perishable Foodstuffs and on the Special Equipment to be used for such Carriage (ATP) 1970 United Nations, New York. E/ECE 810 Rev. 1, E/ECE/TRANS/563 Rev. 1. 9. Industry Guides to Good Hygienic Practice: Baking Guide, Catering Guide, Markets and Fairs Guide, Retail Guide, Wholesale Distributors Guide, Fresh Produce Guide. Chadwick House Group Ltd 1997–99. 10. ISO (International Standards Organisation) 9000 Series of Standards 1994. ISO 9000: Quality Management and Quality Assurance Standards, Part 1: Guidelines for Selection and Use. ISO 9001: Quality Systems – Model for Quality Assurance in Design/Development, Production, Installation and Servicing. ISO 9002: Quality Systems – Model for Quality Assurance in production and Installation. 11. International Institute of Refrigeration, Recommendations for chilled storage of perishable produce. Paris 1979. 12. Shipowners Refrigerated Cargo Research Association, The transport of perishable foodstuffs. 2nd edn, Cambridge 1991. 13. British Refrigeration Association, Testing of food temperatures in retail establishments, 1986. 14. RFIC (Refrigerated Food Industry Confederation), Guide to the Storage & Handling of Frozen Foods, Methyr Tydfil, Stephens & George Ltd, 1994. 15. Institute of Food Science & Technology, Guidelines for the Handling of Chilled Foods 2nd edn, 1990 16. Department of Health, Guidelines on the Food Hygiene (Amendment) Regulations 1990, HMSO, 1991. 17. BS EN 441-5: 1996 Refrigerated display cabinets: Part 5. Temperature test. London, British Standards Institution. 18. BS EN 441-4: 1995 Refrigerated display cabinets: Part 4. General test conditions. London, British Standards Institution. 19. JAMES S J, EVANS J A and STANTON J, Performance of Domestic Refrigerators, Proceedings of 11th International Conference on Home Economics,13–15 Sept., Middlesex Polytechnic, UK 1989. 20. TUCKER G, Guideline No. 1: Guidelines for the use of thermal simulation systems in the chilled food industry. Campden and Chorleywood Food Research Association, 1995. 132 Chilled foods 21. British Standard BS EN 12830:1999, Temperature recorders for the transport, storage and distribution of chilled, frozen and deep-frozen/quick frozen food and ice cream – tests, performance and suitability. London, BSI. 1999. 22. Draft CEN Standard prEN 13485, Thermometers for measuring air and product temperature for the transport, storage and distribution of chilled, frozen and deep-frozen/quick frozen food and ice cream – tests, performance and suitability. Brussels CEN March 1999. 23. Draft CEN Standard prEN 13486, Temperature recorders and thermometers for the transport, storage and distribution of chilled, frozen and deep-frozen/ quick frozen food and ice cream – periodic verification. Brussels CEN March 1999. 24. BS 4937: Part 20. Specifications for thermocouple tolerances. London, British Standards Institution, 1983. 25. BS 1904: 1984. Specification for industrial platinum resistance thermometer sensors. London British Standards Institution, 1984. 26. FAIRHURST D, Temperature monitoring in the cold and chill chain, (A one- day Seminar sponsored by MAFF, 30.01.1990.) Food Science Division Report, MAFF, London. 1990. 27. KLEER J, PASTARI A, WIEGNER J and SINELL H, Recording temperature patterns with modern recording systems (original German), Fleisch- wirtschaft, 1991 71 (6) 698–704. 28. JAMES S J and EVANS J A, ‘The accuracy of non contact temperature measurement of chilled and frozen food’, IChemE Food Engineering Symposium, University of Bath, 19–21 Sept. 1994, Publication No. 106, FPERC, University of Bristol 1994. 29. British Standard BS 7908: 1999 Packaging – temperature and time– temperature indicator– performance specification and reference testing. London, British Standards Institution, 1999. 30. TAOUKIS P S and LABUZA T P, ‘Application of time–temperature indicators as shelf-life monitors of food products’, Journal of Food Science, 1989 54 (4) 783–8. 31. TAOUKIS P S and LABUZA T P, ‘Reliability of time–temperature indicators as food quality monitors under non-isothermal conditions’, Journal of Food Science, 1989 54 (4) 789–92. 32. BALLANTYNE A, An evaluation of time–temperature indicators, Technical Memorandum No. 473, Campden and Chorleywood Food Research Association, 1988. 33. SELMAN, J. D. and BALLANTYNE, A., ‘Time–temperature indicators: Do they work?’, Food Manufacture, 1988 63 (12) 36–8, 49. 34. SELMAN, J. D., ‘Time–temperature indicators: how they work’ Food Manufacture 1990 65 (8) 30–1 and 33–4. 35. Ministry of Agriculture, Fisheries and Food Publication, Time–temperature indicators: Research into consumer attitudes and behaviour, National Consumer Council, 1991. Temperature monitoring and measurement 133 36. GEORGE, R. M. and SHAW, R., A food industry specification for defining the technical standards and procedures for the evaluation of temperature and time–temperature indicators, Technical Manual No. 35, Campden and Chorleywood Food Research Association, 1992. 37. GIGIEL A. J., JAMES S. J. and EVANS J. A., Controlling Temperature During Distribution and Retail, Proceeding of the 3rd Karlsruhe Nutrition Symposium, European Research towards Safer and Better Food,18–20 October 1998, edited by Gaukel V and Speiss W.E.L. pp. 284–92, 1998. 38. FRPERC Newsletter ‘Predicting food temperatures in refrigerated transport’ number 17, May 1997 pp. 4–5, University of Bristol. 39. FOSTER A. M. and QUARINI G. L., ‘Using advanced modelling techniques to reduce the cold spillage from retail display cabinets into supermarket stores’ ICR/IIR Conference, Refrigerated Transport, Storage and Retail Display, Cambridge, 29 March–1 April 1998, FRPERC Publication No. 586, University of Bristol. 134 Chilled foods