Volume 13: Industry Processes and Controls
6. Rendering
Annex B to Chapter 6: Manufacturing process of rendering
(i) Batch systems
(ii) Continuous systems
Temperature and time in batch and continuous systems
Solvent extraction

6.79 This annex describes in more detail the two types of rendering process - batch processing and continuous processing - and discusses the time and temperature commonly achieved in each. It also describes the process of solvent extraction, which took the product of rendering and extracted more tallow from it.

(i) Batch systems

6.80 Up until the 1960s, most rendering plants in the UK used 'dry rendering' (atmospheric) batch processing systems. Before being cooked, a batch of raw material was fed from a storage area into a crusher. In a small plant this might have been done by hoisting the storage container above the crusher and emptying it through a hatch. In a larger, more automated plant it was more likely that an auger would load the material into the crusher. The crusher reduced the particle size of the raw material to about an inch, broke up the fatty and gut tissue and conveyed the resultant mass to a cooker. ¹ The cooker was a large, steam-jacketed vessel. Inside, a revolving beater shaft helped break down the fatty tissue even further. The cooker was heated at normal atmospheric pressure to around 100°C until the moisture was driven off in the form of steam and released through vents. Of the total processing time, the majority was spent in driving off the moisture. Once this was done, the temperature would rise to 140°C or more. ² This increase in temperature would cause the cell structure of the residue to break down, releasing the fat as tallow. In some plants the load was discharged once the maximum temperature was reached; in others there was a holding time of up to 30 minutes. The tallow was then run off into a separate container and the solids emptied from the cooker.

6.81 Another method of batch rendering was 'wet rendering', in which the raw material was subjected to a temperature of 140°C under high pressure generated either by injecting steam into the cooker, or by allowing the steam from moisture in the raw material to build up. ³ The renderer might choose first to raise the temperature to the maximum and hold it for a while, and then slowly release the pressure, sending the temperature back to around 100°C. The exuded tallow could then be run off and purified by gravity or centrifugation to remove any water and particulate matter. The moist solids would then be dried at this temperature for up to three or four hours. Alternatively, some renderers simply cooked the raw material at an increasing temperature for two or three hours before reaching the chosen maximum temperature, whereupon the material was removed (either immediately or after a holding time) and the tallow extracted using a press. ⁴

(ii) Continuous systems

6.82 From the 1970s onwards, a variety of continuous rendering systems became available. They worked in different ways, but all used heating, separation and cooling on a continuous flow basis - essentially, raw material was fed in one end of the cooker and the finished product ejected at the other. ⁵ In most, the heat treatment was similar in principle to the dry rendering batch systems, but more automated. Automation meant that more control was exercised over the crushing of the raw material, and it could be reduced to consistently smaller pieces. Smaller particles meant better and more consistent heat penetration. Automation also meant that the temperature and timing of the cooking process could be better controlled. If the temperature was wrong, the control systems would not allow more material to be fed into the cooker or allow material to be discharged. ⁶

6.83 In a typical continuous rendering system, a continuous supply of raw material was delivered into bunkers in the floor. This was automatically transported into a crusher, which reduced it to pulp. The workings of the heating stage varied according to the type of plant:

1. Stork Duke

   The system works on the principle of a deep fat fryer. The rendering vessel operates with a high proportion of liquid fat, the heat being applied via a steam jacket and a steam-heated tube rotor. The particle size of raw material entering the cooker is 20-50 mm. Maximum temperatures achieved are between 135°C and 145°C with an average residence time of at least 30 minutes. The protein material is then processed before being ground into meat and bone meal.

2. Stord Bartz

   There are a number of different types and sizes of Stord Bartz systems. The material (reduced to a particle size of 20-50 mm) is heated by a steam-heated disc rotor. The discs occupy the length of the rendering vessel. The average maximum temperature achieved is around 125°C and an average residence time of between 22 and 35 minutes. Pressing and grinding [into MBM] is similar to the procedure operated in the previously described system.

3. Anderson Carver-Greenfield

   Finely minced (to a size of less than 10 mm) raw material is first mixed with recycled heated tallow to form a slurry. This is then pumped through a system of tubular heat exchangers with vapour chambers under partial vacuum before being centrifuged and pressed [into MBM]. The heat treatment involves a maximum process temperature of 125°C with an average residence time of between 20 and 25 minutes.

4. Low Temperature Systems - Protec and Stord Bartz De-watering Rendering Process

   Raw material is initially minced to a particle size of 10 mm before being heated to 95°C for 3-7 minutes. The liquid phases (fat and water) are removed by centrifuging or light pressing and are further separated to recover the tallow. The resultant solids are dried at temperatures ranging from 120°C to 130°C. ⁷

6.84 Some types of continuous rendering system cooked the material at a high temperature before the tallow was removed (similar in principle to the dry rendering batch systems). In others, tallow could be removed throughout the process. In the low temperature systems, because the tallow and moisture were extracted at a temperature of 95°C or below, the tallow was subjected to less pummelling and lower heat than it was in other cookers. This meant that the colour was more likely to be acceptable and, therefore, the value of the tallow higher. Also, less energy was used to extract it. The term 'low temperature' referred to the means of producing the tallow, not the temperature to which the solids were heated. ⁸ There is evidence that, during low-temperature rendering, once the tallow had been removed, there was better heat penetration into the solid particles than during the other rendering methods, and thus more chance of destroying heat-sensitive organisms. ⁹

Temperature and time in batch and continuous systems

6.85 The temperature and time of the rendering processes were balanced not only to produce the desired product, but also to achieve 'sterilisation' as required under the Meat (Sterilisation and Staining) Regulations 1982.

6.86 Cooking temperatures and times for both batch and continuous rendering systems varied considerably, depending on what was being rendered. For instance, if the product was high in fat and low in moisture (as edible fat is), tallow in the material would melt out of the solid at around 45-50°C. Once the material reached 100°C, the moisture would be driven off and the solid residue would cook very quickly, virtually frying in the hot tallow. On the other hand, material such as offal, which was higher in moisture and lower in fat, would take much longer at a higher temperature. ¹⁰ This could lead to discoloration and devaluation of the tallow, which was not of so much concern to producers of low-grade tallow. ¹¹ The particle size of the raw material would also have been taken into account. ¹² Most renderers chose maximum temperatures below 140°C, because at that temperature the vitamins and trace elements in the solids were not too much affected, but the solids were sufficiently crisped to make grinding easier. ¹³ Some renderers of low-quality material could afford to cook it at higher temperatures.

6.87 Mr Bacon, a member of UKRA, said that in dry rendering batch systems, after the moisture was removed, the temperature would rise to 140°C or more. ¹⁴ This statement was supported by Dr Gracey, who also said that the temperature was similar in the wet rendering batch systems. ¹⁵ Both Mr Bacon and Dr Gracey were of the opinion that total processing times for dry rendering were usually about three or four hours. ¹⁶ Investigations by MAFF in 1988 found that batch rendering systems operated at maximum temperatures ranging from 102 to 150°C, often taking between an hour and a half and two hours to drive off the moisture before they could begin to reach these temperatures. In some plants the load was discharged as soon as the maximum temperature was reached; in others the temperature was held for up to 30 minutes. None of the plants observed by MAFF was operating under pressure (that is, as wet rendering systems were). ¹⁷

6.88 Until the late 1970s the temperature controls for batch rendering systems were elementary. ¹⁸ The cookers had gauges on the front indicating the temperature in the cooker chamber, the pressure of the steam in the jacket and possibly the pressure in the chamber. Generally temperatures were not recorded or monitored systematically. ¹⁹ Furthermore, although high temperatures were reached, they were very variable and not monitored effectively. With batch rendering, it was not necessarily a temperature, but a quality of product that was to be achieved. ²⁰ In the absence of proper managerial control, it was possible to discharge the material before it was completely cooked. ²¹ This might have happened when the end of the shift was approaching or if there was too much material for the plant to process. For example, it would have been possible to discharge the contents after heating for about two hours at 100°C, which could have been long enough to have driven off the moisture and extracted the tallow, allowing the solid material to be ground into MBM. The definition of 'sterilised' in the 1982 Meat (Sterilisation and Staining) Regulations included 'dry rendered . . . into technical tallow, greases, meals, feeding meals or fertilisers'; so the material in this case would have been 'sterilised' in terms of the Regulations. However, the treatment would not necessarily have been sufficient to render the material sterile in the sense of eliminating all bacteria, viruses, etc. There were, of course, some limitations on the short cuts that could be taken, since the resulting solids would still need to be saleable.

6.89 PDM was one of the renderers that sometimes processed greaves that had been produced by other renderers. Their experience of greaves sourced from other companies also suggested that batch processing was highly variable in terms of time, temperature and product condition. ²²

6.90 In giving evidence in relation to the above, witnesses generally referred to batch rendering in the 1960s and 1970s. However, the results of surveys of rendering plants by Mr Wilesmith's staff, conducted in 1988, indicated that, while some batch rendering plants had the capacity to monitor temperatures effectively by then, and did so, many still did not have the necessary equipment, and appeared not to use temperature as an important guide when rendering. ²³

6.91 As noted above, the greater automation of continuous rendering systems allowed more accurate control over both the time and temperature of rendering. The renderer therefore had greater flexibility, allowing a wider range of time and temperature combinations to be used in cooking the material.

6.92 The Inquiry learnt of a variety of time and temperature combinations that have been used in the different types of continuous rendering systems. For example, the 1988 MAFF investigations mentioned above found the following:

1. Stork Duke cookers were all operated at a maximum temperature of between 135 and 145°C. Because of the design of the cooker it was difficult to determine the time taken for material to pass from one end to the other. One renderer said that 65 minutes was claimed in German studies, but because the material is pushed through the cooker by new material being fed in, the time depends on what rate of throughput is selected by the renderer. The time for which the material is exposed to the maximum temperature is not certain either.
2. Stord Bartz driers operate in a range from 125 to 145°C, except for one plant which operated at 80°C. Again, time spent in the cooker depended on the throughput rate, although experiments indicated that the minimum time was 60 minutes, with an average of 2½ hours. It was believed that the material was exposed to the maximum temperature for only 30 minutes of the total time.
3. Carver-Greenfield systems are believed to process the material over 30 to 40 minutes, operating at a maximum temperature of between 104 and 123°C, depending on the material type. The material was exposed to the maximum temperature for only about 15 minutes.

Protec systems were low-temperature rendering systems operated at two plants, at both of which the material was heated for 3 to 7 minutes at about 95°C before the tallow was extracted. At one plant, the solid residue was then dried in a batch cooker at 120 to 130°C. At the other, the residue was tossed around inside a rotating barrel for 25 minutes while being blasted with air that entered at 700-800°C and exited at about 110°C. However, the temperature of the material inside was unknown.

Solvent extraction

6.93 Sometimes, the greaves produced by renderers were further processed to remove more tallow, in a secondary, separate process known as 'solvent extraction'. Greaves produced using either a batch or continuous rendering system could be subjected to solvent extraction, usually after cooling. However, most continuous rendering systems incorporated an expeller press, removing further tallow by pressure and processing the greaves into MBM directly.

6.94 In a statement, Mr Bill Bacon, a former member of the UKRA Council, said:

. . . the majority of solvent plants in use in the U.K. were Static Pot plants manufactured by Iwel Ltd. These static pots were cylindrical mild steel vessels constructed to contain 3 or 5 tonnes of meat and bone greaves. The greaves were put into the static pot and heated with a steam coil to a temperature of 65-70°C then benzene type solvent was pumped 2 or 3 times into the vessel. The resultant mixture of tallow and solvent being passed through a heat exchanger to vaporise the benzene and recover the tallow, the benzene also being recovered by cooling the vapour down to ambient temperature and subsequently re-used. The material left in the pot was heated to 90°C to vaporise the solvent and stripped using live steam to remove any residual solvent, then discharged for cooling and grinding. As the name implies there was no mechanical agitation in the pot so that when stripping, live steam did force its way through the mass by the weakest channel leaving certain areas of the mass untouched. This meant that the material would never reach a temperature in excess of 90°C. Again although a dangerous practice but occurring due to human frailties, material could be discharged before the full solvent cycle had been completed. ²⁴

6.95 The solvent extraction system described by Mr Bacon is an older style of system which processed greaves in batches. There were other types of system: Mr Foxcroft told the Inquiry that PDM used 'continuous' extraction in one of its plants. In a document on rendering presented to the Lamming Committee by Mr Lawrence, one type of continuous solvent extraction was described as follows:

[The crushed greaves] travels as a counter current to the solvent at a temperature below the boiling point of the solvent. This is carried out at 70°C. This stage of the process takes up to 8 hours. The solvent saturated solids then pass to a steam jacketed desolventiser in which the temperature of the solids is raised to approximately 105°C over a period of 45-60 minutes. The final stage, in the last compartment of the desolventiser, involves the direct injection of low pressure steam to the solids for about 15 minutes to ensure that all the solvent vapour has been removed. ²⁵