A

Direction of convection current flow Fig. 5.18 Relationship between temperature and beer density (adapted from Wilkinson, 1991).

Temperature distribution and fluid flow patterns in large cylindroconical vessels during crash cooling have been studied by Brandon (Larson & Brandon, 1988; Reuther et al., 1995). The studies used a combination of experimental observation in production-scale and model vessels. The ineluctable conclusion was that the mixing patterns are far more complex than is usually appreciated. It was observed that when a beer was being held at a temperature below the inversion point, the predicted upward fluid flow could be detected in beer adjacent to cooling jackets. At the same time, there was also a simultaneous and unexpected downward movement. Similarly, temperature measurements made at a point near the wall indicated rapid fluctuations over a period of a few minutes, indicating the presence of vortex motion. These authors hope for an outcome in which optimum cooling and vessel design may be achieved using predictive models. This seems still to be a distant prospect! Schuch

(1996) concluded that when using a single point thermometer, it was prudent to cover as much of the cone as possible with cooling surfaces and at least 50% of the surface area of the cylinder. Similarly, Manger and Annemuller (1996) reported that it was not possible to make general recommendations regarding the suitability of any one cooling system because of the multiplicity of factors involved. Appropriate cooling systems were best fitted in response to site-specific needs.

Notwithstanding the theoretical and perhaps still unresolved aspects of vessel cooling, some practical points may be made with regard to the optimum placing of cooling jackets for different duties. The much-quoted work of Maule (Maule, 1976; Crabb & Maule, 1978) is invaluable in this respect. In this study, a 1600 hi cylin-droconical fermenter was fitted with a manifold device which allowed temperature measurement and withdrawal of samples at various locations within the vessel. From a study of a number of fermentations, both ale and lager, it was concluded that during primary fermentation the vessel contents were essentially homogeneous throughout. Satisfactory attemperation could be achieved for both lager and ale fermentations, held at temperatures of 8°C using a single wall jacket. This was best located high up the vertical sides of the vessel whereas the thermometer should be placed at the base of the cylinder. This arrangement promotes good convective mixing. Thus, the thermometer is located in the warmest part of the vessel. Output from this activates coolant flow in the jacket above and in response the cooled beer flows downwards due to the increased density. With very tall vessels, especially with high temperature fermentations performed at 21°C, a second jacket was considered advisable.

Crash cool 1 (2°C to 4°C) in a tall vessel required the use of two or more side wall jackets, one located near the top of the cylinder and the other abutting the junction with the cone. In this case the beer was cooled to near the inversion point and to overcome the concomitant lack of mixing, provision of cooling at both high and low level was necessary. In addition, if the beer was held in the same vessel for a long period, a cone-cooling jacket was shown to be necessary to avoid localised heating in this area.

A cone-cooling jacket was shown to be necessary for storage at conditioning temperatures to avoid localised heating of the yeast and beer in the cone. In this case, in the absence of cone cooling, the main body of the beer may be cooled to a temperature lower than the inversion point and thus, it tends to rise. Yeast sediments in the cone and generates metabolic heat, which cannot be dissipated because of the lack of downward convection. The positioning of thermometers was also important. If there was no cone cooling and temperature was controlled by a single thermometer located just above the cone, the rising warm stratum activated the wall cooling jackets leading to overcooling of the main body of the beer. In this case, it was better to have two wall-cooling jackets with the controlling thermometer located mid-way between them. Thermometers should not be located at the top of the vessel because of the possibility that they may be inadvertently exposed to the headspace gas. In this case, over-cooling and ice formation was possible. Thermometer probes must be long enough to project into the main body of liquid, and therefore avoid local cold currents close to the cooling jackets.

Where vessels were used solely for cold conditioning and were filled with beer that had been pre-chilled, satisfactory attemperation could be achieved with cooling jackets located at the base of the vessel. If the beer was delivered warm to the conditioning vessel a second jacket positioned mid-way up the vessel was recommended. In both instances, the controlling thermometer should be placed mid-way up the vertical wall. Obviously, if vessels are used for both primary fermentation and cold conditioning it is necessary to have a combination of wall and cone jackets as well as multiple thermometers in appropriate locations. Such vessels require complex control systems, which activate combinations of cooling jackets, and controlling thermometers, depending on which cooling duty is called upon. The number and positioning of jackets required for cylindroconical vessels with different cooling duties are summarised in Fig. 5.19.

Fig. 5.19 Positioning of cooling jackets and thermometers (T1; T2) for cylindroconical vessels: A, used for primary fermentation only (low-temperature lager-type fermentation); B, used for primary fermentation only (high-temperature ale-type fermentation); C, used for cold conditioning only (centrifuged beer, pre-cooled on receipt); D, used for cold conditioning only (centrifuged beer, not pre-cooled on receipt); E, dualpurpose vessel suitable for all duties including single tank fermentation and conditioning.

Fig. 5.19 Positioning of cooling jackets and thermometers (T1; T2) for cylindroconical vessels: A, used for primary fermentation only (low-temperature lager-type fermentation); B, used for primary fermentation only (high-temperature ale-type fermentation); C, used for cold conditioning only (centrifuged beer, pre-cooled on receipt); D, used for cold conditioning only (centrifuged beer, not pre-cooled on receipt); E, dualpurpose vessel suitable for all duties including single tank fermentation and conditioning.

The sole use of wall cooling jackets, particularly for cold conditioning, has obvious disadvantages because of the lack of mixing. This fundamental drawback can be ameliorated by providing forced agitation in the vessel. Small mechanical agitators, as illustrated in Fig. 5.11, may be fitted for this purpose; however, care must be taken to ensure that the surfaces are cleaned properly by the CiP system. A less invasive approach is to use gas agitation. Ladenburg (1968) reported that injection of liquid carbon dioxide into the base of a cylindroconical fermenter produced sufficient agitation such that the same rate of heat transfer was obtained as with a mechanical turbine impeller. Furthermore, cooling was assisted by evaporation of the liquid carbon dioxide. Similarly, Masschelein and colleagues (Lemer et al., 1991) showed much reduced cooling times to conditioning temperatures in fermenters injected with gaseous carbon dioxide. In this case slow addition rates were used in a gas lift approach, which, it was claimed, produced circulation currents similar to those produced by thermal convection. Despite the attraction of this approach to improving temperature control, it is yet to see widespread adoption. It does have the disadvantage that it may lead to over-carbonation, thereby requiring further adjustment of carbon dioxide levels during subsequent processing.

5.4.3 Dual-purpose vessels

As discussed in the previous section, cylindroconical vessels with steep cones may be used for both primary fermentation and cold conditioning. Although this may be a single tank operation (Harris, 1980), this use is comparatively rare. There are, however, a small number of alternative large-capacity vessels that were designed specifically for combined fermentation and conditioning. These were developed during the 1960s and '70s to cope with the demand, at the time, for cost-effective increased capacity and accelerated throughput.

The single tank concept is contentious and many brewers consider that fermenting and conditioning should be distinct operations performed in dedicated vessels. Dualpurpose vessels are maybe over-engineered since they must cater for all duties. Thus, conditioning vessels are essentially very simple since they need do little more than hold the green beer at a low temperature. The presence of facilities for yeast cropping, carbon dioxide collection, monitoring and control associated with primary fermentation are unnecessary and potential hygiene hazards. Conversely, the cooling capacity of dual-purpose vessels must be sufficient to achieve and hold at sub-zero conditioning temperatures. This degree of cooling is not needed during primary fermentation and indeed the temperature differential between coolant and wort is so great that there is a risk of thermal shock to the yeast. Furthermore, the transfer between separate vessels allows greater flexibility. Thus, there is more opportunity for blending different batches of green beer, in-line additions may be made and in-line cooling is possible.

There are advantages to a single-tank operation, the most important of which is the reduction in total process time, compared to the two-tank approach. There is an element of flexibility in that it is not necessary to decide on what is the ratio of fermenting to conditioning vessels most appropriate to any particular brewery. Instead, calculations may be based simply on total capacity requirements and individual batch size. The single tank approach avoids the intermediate tank-to-tank transfer and therefore avoids the possibility of oxygen pick-up. Losses associated with vessel emptying are halved as are the number of cleans. Maule (1976) reported that cooling to conditioning temperature in a single tank was quicker and resulted in less temperature stratification than chilling of similar beer after transfer. Fricker (1978) considered this was due to beer stratification. In the single tank, convection currents are generated by evolution of carbon dioxide released from deeper saturated strata moving upwards to regions of lower pressure when cooled to below the inversion point. Presumably, during the tank-to-tank transfer the stratification would be destroyed, and, on subsequent cooling, convection current flow would be less dramatic.

In-line chilling during tank-to-tank transfer removes much of the cooling duty of the conditioning tank, as discussed in the previous section. However, it does make large demands on the brewery refrigeration plant. Luckiewicz (1978) described a brewery operation in which beer cooled in fermenter to 5°C was transferred to a conditioning tank via a heat exchanger. Transfer time was 4 hours, during which the beer temperature was reduced to 0°C. It was found that during this operation the refrigeration load for the brewery increased four-fold over all other duties. This necessitated a high capital investment to cope with a relatively transient peak demand. When the transfer time was extended to 8 hours, the differential between the average and peak refrigeration loading was reduced to a much more acceptable level.

Ultimately the choice between the single- or two-tank approach depends on the throughput and batch size requirements of the individual brewery. Undoubtedly the single-tank approach is more suited to large batch sizes and just a few beer qualities. In smaller breweries, or where there is a need to produce short runs of several different beer qualities, the dual-tank approach probably offers more flexibility.

5.4.3.1 Asahi vessels. Asahi vessels, developed by the Japanese brewing company of the same name during the mid-1960s, were the first designed purposely for uni-tank operation (Takayanagi & Harada, 1967). Apart from Japan, Asahi vessels were installed in many United States breweries, also during the 1960s and '70s, where they were seen as a cost-effective method for increasing fermentation and conditioning capacity (Lindsay & Larson, 1975). These authors reported on the installation at the Genesse Brewing Company, New York State, of two Asahi-type vessels, each with a capacity of 10 600 hi. These were two of the largest known fermentation vessels ever to be constructed.

Vessels are made from stainless steel and are cylindrical. The total capacity is 5000 hi and the aspect ratio is close to 1, the internal diameter being 7.5 m and the height 11.8 m. The characteristic feature of the vessel is the internal base, which slopes towards the exit main to facilitate removal of yeast and other solids (Fig. 5.20). Attemperation is provided by jackets in the wall and base. The jackets are surrounded by insulation and an outer weatherproof skin. For single-tank operation additional cooling is obtained by circulating the vessel contents through an external heat exchanger. The circulation loop also includes a centrifuge by which means the yeast loading may be reduced. Circulation of the vessel contents by-passing the centrifuge and chiller is also possible. The point of beer re-entry consists of a stainless steel pipe, which is pivoted. This arrangement allows improved control of the dissolved carbon dioxide concentration.

Operation of the single tank operation was described by Amaha et al. (1977). Prior to filling the vessel, wort cold break was removed using a flotation treatment. All the yeast, which was a weakly flocculent and highly attenuating type, was pitched with the first wort addition. Vessel filling required five brewlengths and this took 20 hours. Fermentation and conditioning was of the traditional lager type, being lengthy and using comparatively low temperatures. The initial temperature was 6°C and during primary fermentation this rose to 9°C then decreased to 5°C. At the end of primary fermentation, which took 8 days, the green beer was pumped via the in-line chiller and centrifuge to both accelerate cooling and reduce the yeast count. This process was continued for 7 hours, with a pumping rate of 350 hi h l, and during this time the temperature fell between 2 and 3°C and the suspended yeast count from c. 55 to c. 25 million cells ml 1. The green beer was then matured using a traditional lagering process for a further period of some 30 days. During this time the temperature was further reduced to — 1°C. Mixing during the lagering period was improved by pumping the beer through the external loop by-passing the chiller and centrifuge. At the end of lagering approximately 80% of the yeast, which was suspended at the end

CiP inlet CO, outlet

CiP inlet CO, outlet

Fig. 5.20 Schematic representation of an Asahi vessel (adapted from Amaha et al., 1977).

of primary fermentation, was shown to have formed a compact sediment that was easily removed from the vessel.

Amaha et al. (1977) reported that based on five years' experience, beer made using a single-tank method in Asahi vessels was indistinguishable from that made by the two-vessel approach. The single-tank method was shown to be the most cost effective. Thus, initial capital and subsequent running costs were reported to be 88% and 65% respectively of those incurred by the conventional two-tank method.

5.4.3.2 The Rainier uni-tank. Like the Asahi vessel the uni-tank also arose during the 1960s, in this instance at the Rainier Brewing Company of Seattle, Washington State, USA. Similarly, they were also designed specifically for single-tank fermentation and conditioning, hence the name which is a contraction of'universal tank' (Fig. 5.21).

The reasoning which led to the design of this vessel is described by Knudsen and Vacano (1972). The optimum tank size, taking into account production requirements and the capacity of the brewhouse, was considered to be 5500 hi. The intention was to enclose this volume within a tank of a geometry that used the minimum quantity of construction materials. A spherical configuration was discounted on the basis of prohibitive construction cost, and therefore a cylinder with an aspect ratio close to 1 was chosen. To maintain the low aspect ratio and again keep costs to a minimum, the vessel was fitted with a shallow conical base with an included angle of 12.5°.

Fig. 5.21 Rainier uni-tank (redrawn from Knudsen & Vacano, 1972).

The vessel was constructed from readily available stainless steel plates, each 1.2 m wide. Using seven courses of these gave a tank with a straight side measurement of 8.5 m and the same internal diameter (8.5 m x 8.5 m). Cooling was provided by liquid ammonia fed to a dimpled jacket mounted towards the top of the vertical wall. Insulation was provided by a 15 cm layer of polyurethane foam, held in place by an outer weatherproof aluminium skin. The coolant capacity was capable of reducing the temperature by c. 5.6°C in 24 hours, starting from 13.3°C. This was shown to be adequate for attemperation during the most active period of the fermentation. Thus, although the aspect ratio of the vessel was chosen to minimise the use of construction materials, it was still apparently capable of providing an adequate cooling area, in this case c. 0.005 m hi Cooling to the cold conditioning temperature was assisted by injection of carbon dioxide. This was achieved by directing the flow of gas upwards through a ring located in the centre of the tank at the base of the vertical wall. In addition to promoting convective cooling, the gas flow assisted in sedimenting the yeast in the cone and stripping unwanted volatile beer components. The exhaust gas was either liquefied and collected or purified and re-introduced into the tank.

When used as a uni-tank the vessel was filled with 10 brewlengths of wort, representing 87% of the total volume. On some occasions, vessels were filled to over 90% of the total capacity. This high-volume utilisation rate was possible because it was claimed that the good mixing characteristics of the uni-tank, presumably a function of the low aspect ratio, resulted in only small generation of foam. Primary fermentation lasted for 3-4 days and was performed at 13°C. After a further 2-3 days' warm conditioning to reduce vicinal diketone levels, the beer was cooled to — 1.7°C over a period of 6 days. The yeast was removed after primary fermentation and was suitable for retention for future pitching. The relatively shallow cone had no adverse effect on yeast cropping, at least with this strain. After beer removal, vessels were cleaned with cold gluconated caustic and hypochlorite via an automatic CiP system.

5.4.3.3 Spheroconical fermenters. Spherical vessels offer many advantages. The sphere is the optimal geometry for enclosing the maximum volume within a minimal surface area, and therefore it is the most cost-effective shape with respect to usage of construction materials. In addition, it is an ideal shape for withstanding pressure. Furthermore, it would be suspected that a spherical fermenting vessel with wall cooling would naturally generate strong convection currents, and thereby facilitate good mixing and attemperation. Lastly, it presents a relatively easily cleaned interior and, compared to other vessels, a minimum surface area. Despite the misgivings of some, for example Knudsen and Vacano (1972), that the construction of spherical vessels would be precluded by the cost and difficulty of construction, others have successfully followed this route to fruition. Thus, the Aguila Brewery in Madrid has designed and installed such vessels to be used in a uni-tank process (Martin et al., 1975; Posada, 1978).

The principal features of the vessel are shown in Fig. 5.22. Since it would be very difficult to crop yeast from an entire sphere, a conical base is fitted. The vessels are constructed from stainless steel, surrounded by 220 mm thick foam insulation and an exterior coating of epoxy resin. Cooling is provided by wall jackets through which a 25% aqueous solution of propyleneglycol at — 4°C is circulated. The jackets are arranged in four rings around the spherical part of the vessel. In addition, there is a cone cooling jacket. The total cooling surface is 150 m2. The capacity of the vessels is 5000 hi. The diameter of the sphere is 10 m and the height, including the cone, is 11.95m.

Martin et al. (1975) described a uni-tank process using spheroconical vessels. The wort had a gravity of 11.4°Plato, the initial dissolved oxygen concentration was 35 ppm, the starting temperature 12°C and the yeast pitching rate 30 x 106 cells ml 1. During primary fermentation, which lasted for 4 days, the temperature rose to 14°C. After this time, the green beer was cooled to 8°C, over a period of 20 hours. The yeast was a highly flocculent type, which facilitated sedimentation in the cone such that, unlike the Rainier uni-tank, it was not necessary to use a centrifuge to reduce the cell count at the end of primary fermentation. After removal of the yeast crop, the beer was further cooled to 0°C and held for 21 days for maturation. During cold conditioning, the carbonation was adjusted to 5.5 gl l. This process was assisted by the ability to pressurise the vessel. It was noted that carbonation levels at the end of cold conditioning were, respectively, 5.3 and 5.7gl 1 at the top and bottom of the vessel. This relative homogeneity was taken as evidence of the excellent mixing characteristics of spheroconical vessels.

As with the Rainier uni-tank, beers produced in spheroconicals using a single-tank process were indistinguishable from similar beers produced in conventional fermenters and cellar tanks. However, it was observed that it was necessary to reduce hop

Fig. 5.22 A spheroconical fermenter.

rates by 12% in the uni-tank process. This improved utilisation was thought to be due to a reduction in the loss of bittering components on the yeast as a result of reduced foaming in the spheroconical vessel.

Brew Your Own Beer

Brew Your Own Beer

Discover How To Become Your Own Brew Master, With Brew Your Own Beer. It takes more than a recipe to make a great beer. Just using the right ingredients doesn't mean your beer will taste like it was meant to. Most of the time it’s the way a beer is made and served that makes it either an exceptional beer or one that gets dumped into the nearest flower pot.

Get My Free Ebook


Post a comment