Structure Of Fermenter

^Ti = Time to racking gravity (h). *T2 = Warm conditioning time (h).

height restriction. The majority of services are supplied to the base of vertical vessels in a simpler arrangement, which uses fewer valves. With very tall vertical vessels, the large hydrostatic head requires heavy-duty pumps to fill from the base. In this regard, horizontal vessels present less of a problem; however, even this advantage is lost where the vessels are stacked.

On balance, vertical cylindrical vessels are more advantageous than their hori zontal counterparts and this is reflected in their relative use. As early as 1984, Shuttlewood reported that, of all new cylindrical vessel installations, less than 1 in 20 were of the horizontal type. The move towards vertical geometry seems set to continue.

5.4.2 Cylindroconical fermenters

This style of vessel was originally designed by Nathan (1930a, b) at the end of the nineteenth and introduced by him, initially to Switzerland, as a means of improving the efficiency of the traditional lager fermentation. After this time they slowly came into common usage in mainland Europe and throughout much of the rest of the world. The United Kingdom resisted wholesale adoption of cylindroconicals because of the belief that they were unsuitable for the production of top- fermented ales. This prejudice has now been largely dispelled (see Section 5.1.4) and cylindroconicals are now the most widely used high-capacity fermenters in the United Kingdom and elsewhere.

Installation of cylindroconical vessels (CCVs) was particularly prevalent during the 1960s and '70s. During this period, in many countries, there was a trend towards fewer and larger brewing companies, producing brands for both national and international markets. This produced a requirement for larger batch sizes. Simultaneously, the fabrication and material costs of large stainless steel vessels fell and this coincidence of demand and economic feasibility paved the way for the introduction of cylindroconicals and other related high-capacity vessel types. Since that time little has changed in terms of vessel design. Recent advances mainly relate to improved methods for monitoring and control and these are discussed in Chapter 6.

The original Nathan process improved the efficiency of lager production by combining primary fermentation and cold conditioning in a single tank. In modern parlance, this was a 'unitank' operation. Modern cylindroconical vessels are essentially the same as the original Nathan design, and, similarly, they may be used for both primary fermentation and cold conditioning. However, in the majority of breweries these operations are separate and involve an intermediate tank-to-tank transfer. Versions with a slightly simpler design may be used solely for primary fermentation, others just for cold conditioning. An idealised representation is given in Fig. 5.11 together with photographs of various CCVs in Fig. 5.12. The vessel as depicted would be suitable for carrying out both primary fermentation and cold conditioning and it shows some fittings, which may be considered as optional. The daunting challenge of delivery and installation is captured in Fig. 5.13.

The original Nathan vessels were constructed from aluminium. This was a disadvantage in that the propensity of this metal to corrode was not fully understood and this did little for their initial reputation. Subsequently, mild steel lined with epoxy resin was commonly used but this has now been superseded largely by stainless steel. The essential feature of the vessel is the replacement of the lower dished end by a cone. The interior surface of this is highly polished to reduce friction and thereby facilitate the collection and removal of the yeast crop. The vessels may be located within a building or outdoors. If the latter option is chosen, external weatherproofing must be provided. Frequently, the fermenters are grouped such that the cones and associated

Cylindoconical Fermenter
Fig. 5.11 A cylindroconical fermenter.

services are enclosed within a hygienically designed room. The ceiling of the room provides mechanical support for the vessels and access to the area at the base of the cylindrical portion, a convenient location for a sample cock. In one reported instance an installation was made in which the vessels were uninsulated and erected within a refrigerated room. The latter provided attemperation during primary fermentation. Circulation of the green beer through external heat exchangers provided a means of cooling to cold conditioning temperatures (de Witt & Hewlett, 1974).

In the majority of installations, attemperation is via wall cooling jackets. These are surrounded by an insulated outer skin to minimise heat pick-up from the environment. Thermometers (T1; T2, Fig. 5.11) are provided within the vessel and output from these is used to control temperature automatically via the supply of coolant to the wall jackets. The number and location of the thermometers are dependent on the duties to which the vessel is put. The natural agitation of fermentation may be augmented by the provision of a small mechanical agitator. Occasionally, a loop system may be used, where the vessel contents are circulated out from the base and back into an entry point at the top. The latter system may also incorporate an in-line heat

Structure Breweries
Fig. 5.12 Cylindoconical fermenters (a) courtesy of Briggs PLC. (b) kindly supplied by Maxine Bellfield of Vaux Breweries, (c) kindly supplied by Harry White. Bass Brewers and (d) courtesy of Ian Dobbs. Bass Brewers.
Colin Turton
Fig. 5.13 Delivery (a), lifting (b) and installation (c) of new CCVs to Bass Brewers Burton Brewery (kindly supplied by Colin Turton, Bass Brewers).

exchanger to supplement or even replace the wall cooling jackets. In addition, postcollection aeration/oxygenation systems can be installed which will also facilitate mixing. Vessels are gauged to allow an accurate measure of wort volume and where appropriate, provide a means of calculating the rate of addition of dilution liquor. In some installations the traditional manual sight glass or dip plate arrangement is replaced by an in-line flow meter located up-stream of the fermenters. Other refinements in monitoring and control of fermentations have been introduced in the last five years and these are discussed in Chapter 6. Vessels are suitable for collection of evolved carbon dioxide and may be pressurised if designed to the appropriate specification.

Filling and emptying of vessels is via a single entry point at the base of the cone. This simplifies the design of the vessels but requires a complex valve block so that the process flow can be directed to the appropriate destination. The pipework is arranged so that a number of vessels may be serviced by a minimum number of common mains (Figs 5.14 and 5.15). More complex valving and duplication of service mains, although providing greater operational flexibility, will, of course, escalate capital and revenue costs.

Wort in

Wort in

Vaux Breweries
Fig. 5.15 CCV cones and associated mains (kindly supplied by Maxine Bellfield of Vaux Breweries).

In operation the vessels and associated mains are cleaned and sterilised automatically by a dedicated CiP system. Wort is delivered to the vessels already oxygenated and inoculated with yeast. Both of the latter operations are performed in-line and may be controlled automatically (Chapter 6). Frequently, the wort may be diluted in-line with brewing liquor using an automatic blending system, to achieve a desired gravity. Alternatively, dilution liquor may be added to the partially filled vessel. If the latter option is chosen, it is important to ensure that the vessel contents are mixed to homogeneity. As shown in Fig. 5.11, this may be achieved using a mechanical stirrer or by gas rousing. Of these two options, mechanical stirring is preferable. With gas rousing there is the inevitable risk of perturbing the initial wort dissolved oxygen concentration, and thereby influencing, in an uncontrolled manner, the outcome of the fermentation (Chapter 6). Thus, the initial oxygen concentration may be supplemented if compressed air is used, or reduced by gas stripping oxygen if nitrogen or carbon dioxide is used.

At the end of fermentation the yeast sediments in the cone from where it is cropped in the form of a concentrated slurry and usually is transferred to a storage vessel. Here it may be retained and used to seed a subsequent fermentation; alternatively, it may be disposed of after first recovering the entrained beer. If the fermenter is used solely for primary fermentation, the green beer is transferred to the next stage of processing. Most of the suspended yeast is removed by centrifugation and cold conditioning of the partially clarified beer is performed in a second chilled (frequently cylin-droconical) storage vessel. Alternatively, with a single tank operation both primary fermentation and cold conditioning are performed in a single vessel. In this mode of operation, the green beer is chilled to between 2 and 4°C at the end of primary fermentation (and possibly a period of warm conditioning) to sediment the yeast. After the yeast crop has been removed, the vessel contents are chilled to a 'conditioning' temperature (typically just below 0°C), and held for an appropriate period to allow flavour maturation and achievement of colloidal stability.

The duties for which cylindroconical vessels are used influences their design. Those used solely for primary fermentation tend to be of a high aspect ratio and have a steep sided cone. Thus, Shuttlewood (1984) reported that the most commonly supplied cylindroconical fermentation vessel is one of approximately 1500 hi with 12-13 m body height, a diameter of 4.0-4.2m and cone with an included angle of 70°. This design favours good mixing and the relatively steep-angled cone facilitates efficient removal of the yeast crop. Similarly, these vessels have good draining properties. Lager fermentations generally make more efficient use of the vessel capacity compared to ales because the latter are usually performed at higher temperatures and this promotes greater fobbing. Consequently, it is necessary to leave a greater headspace (freeboard) to accommodate the foam. This problem can be ameliorated by sparing application of silicone antifoam or other surface active lipids, such as sorbitan monolaurate and glyceryl monoleate (Button & Wren, 1978). Preferably, the addition should be at vessel collection although antifoam can be sprayed directly onto the developing foam. Some brewers view the use of chemical antifoams with suspicion because of the fear of adversely affecting head retention of the finished beer. Mechanical foam breaking devices have not found much favour because they introduce components that are difficult to clean. A new and patented approach, at present only tested at pilot scale, suppresses foaming by the application of ultrasound radiation (Freeman et al., 1997). Whether or not this method sees widespread adoption will require developments in the technology and reductions in cost.

Vessels used for both primary fermentation and cold conditioning tend to have higher capacities, smaller aspect ratios and frequently a much shallower cone. These differences relate to costs versus capacity. Where a single tank process is operated, and therefore fewer vessels in total are needed, it is sensible to use the largest possible batch size, which is consistent with production requirements and the capabilities of the brewhouse. With a single large tank the material costs decrease in relation to the aspect ratio, and furthermore shallow cones are less expensive than the steep variety. On the other hand, a lower aspect ratio reduces the surface area available for wall cooling and shallow cones are less efficient for yeast crop removal and increase wetting losses. The cooling jackets assist in giving structural strength to vessels. Those used for brewing vessels are most commonly of dimpled or half-pipe design, depending on the type of coolant used (Shuttlewood, 1984). The former are used with isothermal refrigerants such as ammonia, whereas the latter are suitable for non-isothermal coolants such as propylene glycol (Wilkinson, 1991). The relative merits of each type of coolant have been discussed by Gerlach (1995). This author considered that ammonia was the most advantageous for several reasons: no glycol transformation stage, higher temperatures at the compressor, smaller pumps, an overall energy saving of 30-40%, smaller refrigeration ducts, more precise temperature control and better flexibility.

Single- and dual-purpose vessels must cater for different cooling requirements and this is reflected in the number and location of wall jackets.

The cooling duties are:

(1) attemperation during the period of maximum exothermy (primary fermentation);

(2) rapid cooling ('crash cool 1') to 2-4°C at the end of primary fermentation and attemperation at this temperature, to sediment yeast;

(3) cooling ('crash cool 2') to the conditioning temperature; and

(4) attemperation at the conditioning temperature.

Several factors determine how much cooling must be provided to fulfil the individual duties outlined above. A proper consideration of these is essential in order to ensure that cooling is efficient and cost effective.

All four duties are influenced by the geometry of the vessel, in particular, the aspect ratio, the capacity, the number, disposition and surface area of cooling jackets, the temperature and flow rate of the coolant. In addition, the thermal conductivity of the material used for vessel construction, the efficiency of the insulation and the temperature of the environment. During primary fermentation, the magnitude of exothermy is governed by the factors which control the rate and extent of yeast growth, principally wort composition, initial dissolved oxygen tension, yeast pitching rate and the required holding temperature. For crash cooling (both 1 and 2) the total temperature drop and the minimum acceptable time to achieve it are influential. With a two-tank arrangement, reduction of green beer to the cold conditioning temperature (crash cool 2) may be accomplished by in-line chilling during transfer. In this case, the second vessel requires only to be capable of maintaining the conditioning temperature. Conversely, with a single vessel operation, or where the beer is transferred without in-line chilling, the cooling capacity must be sufficient to reduce the beer to the conditioning temperature and then maintain it.

Transfer of heat from the beer to the coolant involves a combination of convection and conduction. This can be quantified using the heat transfer formula:

where Q is the rate of heat transfer (Joules s * U is the overall heat transfer coefficient (Jnrs 1K * A is the area of jacket (m2); AT is the mean difference in temperature between the two liquids (K).

In addition to the transfer of heat from the beer to the coolant, there is also the possibility of heat transfer from the environment to the coolant and the magnitude of this effect will also be controlled by the heat transfer formula. For efficient cooling, vessels should be constructed to minimise the value of U with respect to heat exchange between the exterior and the coolant and to maximise the value of U with respect to heat exchange between the coolant and the beer.

Several processes combine to determine these values of U and these are illustrated in Fig. 5.16. Each phase may be viewed as a barrier to heat transfer. Thus, solid materials such as the tank wall, the outer jacket wall, the insulation layer and the outer skin have defined coefficients of thermal conduction. The values of these may be increased by surface soiling (fouling layers). Transfer of heat through the fluid layers is governed by convection. Fluids flowing over plane solid surfaces have a surface film, or boundary layer, where flow may be laminar.

Temperature

Temperature

Exterior Cooling jacket Interior b, OS I J. b2 C b3 f2 TW f3 b4 PF

Exterior Cooling jacket Interior

Fig. 5.16 Schematic representation of the barriers to heat transfer between the exterior and interior of a jacketed and insulated fermenter:A, external atmosphere; bi, air boundary layer; OS. vessel outer skin; I. insulation; I^ outer jacket wall; f^ outer jacket fouling layer; C. coolant; f3. inner coolant fouling layer; TW. tank wall; f3 beer fouling layer; b4. beer boundary layer; PF. process fluid (beer/wort); b2 outer coolant boundary layer; b3. inner coolant boundary layer.

Surrounding this is a phase where flow is turbulent. The size of the boundary layer is influenced by the magnitude of the convection currents in the bulk of the fluid. Where two bodies of fluid of different temperature are separated by a solid layer, the biggest temperature differential occurs within the boundary layers since this region offers the most resistance to heat transfer.

In the case of a fermenter, the insulation should ensure that loss of cooling efficiency due to heat transfer from the environment to the coolant is minimal. This may not always be so. Shuttlewood (1984) estimated that up to 50% of the cooling capacity of fermenters could be dissipated in this way.

Most cylindroconical vessels are made from either stainless or mild steel and both of these have high coefficients of thermal conductivity. However, the metal thickness is governed by mechanical strength and therefore is not a variable with respect to cooling. In a properly designed and cleaned vessel, fouling layers both within the vessel and cooling jackets should be insignificant. The coolant temperature should be sufficiently low so that it is possible to achieve and control the lowest required beer temperature but with no ice formation. Thus, use of very low temperature coolant is an attractive means of increasing the overall heat transfer coefficient but in a well-insulated vessel the inevitable result will be that the beer will freeze. Formation of ice on the inner tank wall dramatically reduces the efficiency of heat transfer. If the coolant temperature is too high, the desired crash cool temperature will either never be achieved or not achieved within an acceptable time span. Coolant is pumped through the jackets and so the fluid flow rate is relatively high, therefore convection is forced and rates of heat transfer are high. The most significant factors which influence heat transfer (or increase the value of U) are the total coolant jacket surface area and the rates of convection within the vessel.

Patterns of mixing within cylindroconical vessels are complex and still not well understood. They vary throughout different phases of fermentation. They are influenced by the number and disposition of the cooling jackets, the aspect ratio of the vessel and the provision, if any, of artificial means of agitation. For this reason, actual values for U are rarely calculated and cooling capacity tends to be provided on an empirical basis. Understandably, most installations are over-provided with cooling capacity. This is wasteful but better than under-estimating cooling requirement. The degree of empiricism is illustrated by the observation of Knudsen (1978). In a survey of the brewing literature at that time he discovered a 100-fold variation in heat exchange area provided per unit volume of vessel capacity (0.006 m2 hi 1 to 6 m2 hi *).

The patterns of mixing, and how they relate to cooling in large vessels, have been investigated and reported. This has been vital in providing information to others to ensure that cooling jackets and thermometers are placed in appropriate locations. During primary fermentation, cooling at the vessel wall causes beer to sink. In response, warmer beer in the central core of the vessel rises and this is aided by the ascending stream of carbon dioxide arising from the bottom of the cone. This pattern of mixing is illustrated in Fig. 5.2. and is based on the observations of Ladenburg (1968). In a later report, Knudsen (1978) claimed that these observations were invalid since they were obtained from a model which used relatively thin wafers to observe patterns of agitation. These failed to take into account boundary layer effects which can obscure what is happening in the deeper recesses of vessels. In addition, this author cautioned that observations made with small-scale model systems do not always reflect faithfully what happens in larger tanks. Knudsen's approach, also using a scaled-down model, was to study hydraulic movements in saline solutions (to simulate wort) containing suspended organic solvent droplets, of the same density as the saline and stained with the lipophilic dye Sudan Black. Agitation was provided by sparging carbon dioxide from the base of the cone of the vessel.

The observed agitation patterns were as shown in Fig. 5.17. Two zones of mixing

Sparge Cone
(after Knudsen, 1978).

were distinguished and flow rates were always more rapid in the upper zone. Upward movement of the lower zone occurred at a velocity proportional to the size of the upper zone. Increasing the sparge rate increased the size of the lower zone but produced a smaller more vigorously mixed upper zone. Decreasing the vessel diameter had the same effect as increasing the sparge rate. The implications were considered that narrow vessels of high aspect ratio would be more prone to foaming, due to the high rates of turbulence in the upper zone. Cooling jackets for attemperation during primary fermentation should be located at the upper zone to maximise free convec-tional heat transfer. More recently, Schuch and Denk (1996) described another model cylindroconical system in which flow patterns were visualised using a laser beam arrangement. The results were similar to those of Knudsen (1978) but there was evidence of much greater complexity of fluid flow.

During the active phase of primary fermentation, there is usually sufficient agitation to ensure homogeneity of vessel contents, and therefore convectional heat exchange is efficient, attemperation is accurate and temperature readings measured at any location are valid. During crash cooling the situation is less satisfactory. At the end of primary fermentation, mixing due to carbon dioxide generation is minimal. In the absence of any other source of agitation, heat transfer is dependent on mixing due to convection currents generated by the formation of density gradients in beers of differing temperature. Weissler (1965) showed that for any beer there is a specific temperature where the density is maximal. As beer is cooled the density increases and hence it tends to sink. At temperatures less than that of maximum density (the inversion point) beer density decreases and the direction of convection flow is reversed (Fig. 5.18). This author demonstrated that the temperature of maximum beer density (TMD) can be calculated from the formula, TMD = 4 — (0.65 RE — 0.24A), where RE is the real extract in "Plato and A is the percent ethanol concentration by weight. For most beers the temperature of maximum density is in the range 2 to 4°C, the same as that used for sedimenting yeast after primary fermentation. The implication of this is that it is possible to have beers of different temperatures but the same density, and therefore stratification is possible if cooling relies solely on convection currents.

tmd tmd

Micro Brewing Process Diagram
Temperature (°C)
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