Cheese can be made from the milk of many mammals including goats, sheep, buffalo, reindeer, camel, llama, zebra and yak. The milk of ruminants is the best milk for cheese making because it contains high levels of the milk protein casein which is required to provide an adequate coagulum. Our consideration of milk composition will include only a summary of the proximate analyses of the most common dairy species and a few, relevant with respect to cheese making, comments about each component.

Gross composition of food (also referred to as proximate analysis) means distribution of the total amounts of fats, proteins, carbohydrates, ash (mainly minerals such as calcium) and moisture or total solids. Typical proximate analysis profiles for cows', sheeps' and goats' milk are listed in Table 4.1. Further discussion refers only to cows' milk unless otherwise stated.

Fat content ranges from 2.0 to 7.0 kg/hl. An approximate average for regions where Holstein Fresian cattle predominate is about 3.9 kg/hl. With respect to cheese manufacture and quality the following properties are important.

Total milk protein ranges from about 2.5 to 5.5 kg/hl. The average for regions in which Holstein Fresians predominate is about 3.3 kg/hl. There are two major groups of proteins, the caseins (about 2.6 kg/hl) which I refer to as the 'cheese proteins' and the whey proteins (about 0.7 kg/hl) which as the name suggests are usually lost in the whey during cheese making. Caseins are not water soluble and so the cow packages them in water dispersable particles called micelles which along with caseins include most of the milk calcium, magnesium, phosphate and citrate (more about casein micelles in Chapter 8). Unlike whey proteins which are very sensitive to heat, caseins are little affected by heating except that they react with heat denatured whey proteins. Table 4.2 lists the principal caseins and some properties which are most relevant to cheese making. Similarly, Table 4.3 lists some properties of the principal whey proteins.

Cheese making principles are similar for milk of all species with some modifications required to account for high solids of some species such as buffalo and sheep. Cows' milk and goat's milk have similar cheese making properties except that:

Through out the modern history of dairying, farmers have selectively bred dairy cattle to increase production or fat content or both. Recently, genetic selection has focussed on other milk properties such as increasing the proportion of milk protein to fat. Three genetic effects are most relevant to cheese making.

Fat content and protein content generally increase or decrease in parallel, but fat varies more with feed and season then protein. The same is true for breed (genetic) effects, such that genetic selection has produced the the following practical effects in modern dairying.

With respect to other solids, mineral content (mainly Ca, Mg, and P) generally varies in proportion to protein content and lactose content is relatively stable. Because lactose is largely a wasted component, increasing protein and fat by feed or genetic selection has economic advantages in terms of feed conversion, milk transportation costs, and waste handling.

Fat content tends to increase during lactation as milk production decreases. The result is that the relative proportion of protein to fat (protein/fat ratio or P/F) is highest at the peak of lactation (about 60 days of lactation) and lowest at the end of lactation. Protein distribution also changes during lactation with resulting effects on cheese ripening and flavour. In particular, the proportion of alpha-caseins decrease during lactation while the proportion of beta-casein increases.

Depending on the relative demand for butter fat versus milk non-fat solids, there may be incentive to change the relative proportions of milk protein and fat. The only short term means to do this is by changing the diet. Generally less roughage and more high energy feeds will encourage lower fat content with little decrease in protein content to provide a higher P/F ratio.

Seasonal variation in milk composition is most important to cheese yield efficiency and composition control. Some important seasonal effects are listed below and illustrated in Figure 4.1. These observations are based on Ontario data.

Cheese making depends on the growth of bacteria to produce acidity, flavour compounds, and ripening enzymes. It is, therefore, important to understand the characteristics of milk as a growth medium.

Milk is a good source of all principal nutrients, including carbon, nitrogen and macro-minerals. Many micronutrients such as vitamins and micro-minerals are also available. However, milk is unique with respect to its sugar.

Carbohydrates, especially simple sugars such as sucrose (table sugar), can be utilized as sources of energy more quickly than fats and proteins. However, the energy currency of the cell is glucose (also called dextrose) so to use available carbohydrates, microorganisms must be able to convert them to glucose.

The only sugar naturally present is milk is lactose. Most microorganisms lack the enzyme lactase which is required to break lactose into its two component sugars, namely, glucose and galactose. Lactic acid bacteria which do have lactase readily break down lactose and use glucose as an energy source. Lactic acid bacteria, therefore, have a competitive advantage in milk; that is, they are able to out grow other bacteria which are unable to obtain glucose from lactose. Further, some lactic acid bacteria are able to convert galactose to glucose.

Acidity as measured by pH is one of the most critical parameters with respect to both food safety and both process and quality control of fermented foods such as cheese. The concepts of acidity and pH are explained in Sections 3.5. The titratable acidity of milk typically varies from 0.12 to 0.19% lactic acid depending on composition, especially protein content. The pH of milk is near the physiological pH of 6.8 which, considering the following points, means that milk is a good growth medium with respect to acidity (pH).

Milk has a high moisture content (typically 87% for cows' milk) and with respect to available moisture, is an excellent growth medium. But, it must be understood that with respect to microbial growth, the critical parameter is water activity not moisture content. Water activity (a_w) is an index of the availability of water for microbial growth. It is the availability of water in the food reported as a fraction of the availability of water from pure water. In other words, the a_w of water is 1 and the a_w of other substances is reported as decimal fractions of 1. Water activity is reduced by dissolved substances, varying directly with number of dissolved molecules rather than the weight. For this reason, relative to large molecules such as proteins, small molecules such as sugar and salt have a large effect on water activity. For example, jams are preserved by their high sugar content.

Microorganisms vary greatly in their ability to survive and/or grow at reduced water activity. However, acknowledging that exceptions exist, the minimum water activity for the principal groups of microorganisms are as follows:

Compare these values with typical a_w values for milk, cheese and a few other foods.

Typical a_w values for some cheese at the marketing stage are given below (Eck and Gillis, 2000). See also typical a_w values for cheese families in Table 1.1.

Moulds require oxygen, so they can be eliminated by vacuum or gas flush packaging. Most yeast are aerobic (require oxygen) but some can grow anaerobically (in the absence of oxygen). Bacteria may fall into any of these categories, but lactic acid bacteria are micoaerophilic or anaerobic.

Milk will acquire some dissolved oxygen during milking, storage and handling, but it is used up quickly during bacterial growth.

The numbered list below identifies seven types of bacteria according to how they change the properties of milk. Often these changes are negative (spoilage) but as we will see in later sections, many of these bacteria are important to the development of cheese flavour. Before proceeding to the list, please note the following definitions:

Keeping the above definitions in mind, note the following types of microorganisms, grouped according to their impact on milk quality.

(1) Lactic acid bacteria which ferment lactose to lactic acid and other end products. Lactic acid bacteria (LAB) important to cheese making will be described further in Chapter 7. For now note the following:

(2) Proteolytic bacteria which degrade protein and cause bitterness and putrefaction. Most important in cheese milk are species of:

(3) Lipolytic bacteria which degrade fats and produce lipolytic rancidity. Again, the most common example in milk is the genus Pseudomonas. Several psychrotrophic species of Pseudomonas produce heat stable lipases as well as proteases.

(4) Gas producing microorganisms which cause cheese openness, floating curd in cottage cheese, and gassy milk.

(5) Ropy bacteria cause stringy milk due to excretion of gummy polysaccharides. Usually ropy bacteria such as Alcaligenes viscolactis are undesirable. However, in some fermented dairy products, ropy lactic acid bacteria such as certain subspecies of Lactococcus lactis are used to develop texture.

(6) Sweet curdling bacteria produce rennet-like enzymes which may coagulate milk. Common examples are the psychrotrophic spore formers Bacillus subtilis and Bacillus cereus.

(7) Numerous off flavours have been associated with specific milk contaminates. Some examples are:

This short course makes no attempt to provide comprehensive training on food safety with respect to cheese manufacture. However, some food safety principles will be discussed in the context of other topics, for example, acid control and food plant sanitation. Here, we mention only some characteristics of a few pathogens which are particularly significant to cheese making. We begin with definitions to distinguish between food infection and food intoxication.

Food infections are caused by organisms which grow in the gastro intestinal track. Illness occurs after ingestion of an infectious dose which depends on many factors including the health status of the person.

Food intoxication results from toxins produced by bacteria. Toxins may be present within the bacteria (endotoxin) or excreted outside the bacteria (exotoxin). The organism need not be alive or even present to cause illness. A good example, is Staphylococcus aureus. Like all the other pathogenic bacteria listed below, Staphylococcus aureus is destroyed by pasteurization but its enterotoxin survives pasteurization.

Lactic cultures are very sensitive to antibiotics. In most jurisdictions increasing penalties have greatly reduced antibiotic residues in milk. Nevertheless, antibiotic testing of all cheese milk is still recommended. See rapid screening tests in Section 3.10.

Mastitis is an infection of the udder which negatively impacts milk quality. Pooling milk dilutes the effect of single infected cows and herds but in most jurisdictions the cumulative effect of mastitis, especially subclinical mastitis, is significant. Olson as cited in Eck and Gillis (2000) estimates a cheese yield loss of 1% if 10% of the milk is from cows with subclinical mastitis. Further, as noted below, the quality effects of mastitic milk are probably of more economic importance than the yield effects.

Causative organisms include human pathogens such as E. coli and Staphylococcus aureus. Nonbacterial infections such as prototheca infection also cause high SCC. Based on new automated procedures for bacteria counting, Ontario producer milk data suggests that prototheca is a common mastitic agent and frequently contributes to high SCC and bacterial counts.

Somatic cells include any type of 'body' cell in the milk, such as skin cells (epithelial) from the cows' udders and leucocytes of several types. Leucocytes are white blood cells which are part of the cow's immune response to infection in the udder, so they are used as an index of mastitis or udder infection. Several observations are relevant:

There is evidence that counts as low as 100,000 cells/ml affect cheese yield (Barbano et al, 1991, J. Dairy Sci. 74:369) and the quality of other dairy products such as ultra-high temperature milk. SCC in the range of 250,000 - 500,000 are associated with altered milk composition and decreased cheese yield. When counts exceed 1,000,000 cells/ml, altered milk composition and reduced cheese yield, are obvious.

Gross composition effects of udder infection are not significant for SCC less than about 250,000/ml. Above that level the following trends are observed:

High SCC are normally associated with shedding of pathogenic (to humans) bacteria in the milk including E. coli, S. aureus and others. Basically, whatever organism is causing the udder infection, including the algae, prototheca will be present in the milk. Further, growth factors present in high SCC milk encourage growth of both E. coli and S. aureus (Amer. J. Vet. Res. 45:2504). The growth rates of some lactic cultures are also affected; Streptococcus thermophilis grows faster and Lactobacillus acidophilus is inhibited.

Perhaps more important than yield are the effects of subclinical mastitis on cheese quality (J. Dairy Res. 53:645). Modest levels of SCC cause several quality problems:

Very high counts (>2 million) will cause milk to taste salty and result in many quality problems. Lower counts, even as low as 300,000 can increase development of bitter flavour due to increased levels of plasmin. This is a particular problem with ultra-high-temperature processed milk because the enzyme is heat stable and the storage time is long enough to permit significant protein degradation.

The following list is a summary of the most important raw milk quality tests. Procedures for some milk quality tests are described in Chapter 3.

Table 4.1 Typical gross composition (kg/100kg) of cow, dairy sheep and goat milk (Wong et al. 1988).

FIGURE 4.1 Seasonal variation of fat, protein, lactose and protein:fat ratio in Ontario producer milk

Clarification may be as simple as filtering out debris or may include standardization of micro flora by removing microbial cells and spores. The principal clarification/standardization procedures are as follows.

(1) Cloth filters are common to remove debris at the farm but should not be necessary at the processing plant.

(2) Centrifugal clarifiers, medium speed centrifuges, remove particles which escape filtration. Cream separators effectively double as centrifugal clarifiers because small particles of debris collect at the periphery of the separator bowl and are ejected as sludge. The loss of milk solids by this process is minimal.

(3) Bactofugation is a high speed centrifugal process which separates bacterial cells and spores. This process is particularly important in Europe where problems arise due to spore formers such as Clostridium tyrobutyricum.

(4) Microfiltration is a membrane process which has been used in a few European cheese plants since 1985. Think of microfiltration as an ultrafine sieve. Microfiltration and related membrane processes are illustrated in Figures 5.1 and 5.2 and are further described in Chapter 22. Microfiltration achieves about 99% reduction of spore forming bacteria relative to 95% by bactofugation. The disadvantage is that microfiltration can be applied only to skim milk because the milk fat globules are too large to pass through the microfiltration membrane (See Figures 5.2).

In addition to standardization of microflora, it is normally necessary to adjust milk fat or protein or both. The objective of milk composition standardization is to obtain the maximum economic return from the milk components. In practice, this means that milk composition is adjusted to achieve the most economically favourable balance of the cost of ingredients and the percent transfer of milk solid components to cheese while maintaining cheese quality.

Cheese yield is mainly determined by the recoveries of protein and fat in the cheese (that is the percent of fat and protein transferred from milk to cheese) and by cheese moisture, but other components also contribute significantly. Cheese yield is discussed in Chapter 12. Chapter 6 is a detailed practical guide to milk standardization, including the necessary calculations for manual standardization. Here we summarize general considerations on milk standardization.

Food regulatory agencies in many jurisdictions have mandated standardized foods for which specific criteria with respect to composition and/or quality must be met. Section 28 Table Part 1, Canada Agricultural Products Act and Regulations lists maximum moisture and minimum fat levels (percent by weight) for 46 cheese varieties. No other composition or quality standards are prescribed, so, the identities of cheese varieties are not protected. For example, American mozzarella is NOT pasta filata cheese like Italian stretch mozzarella, but it is mozzarella according to Canadian regulations.

Table 6.1 includes data for target fat and moisture content according to the respective minimum and maximum values as prescribed by the Canada Agricultural Products Act. It also includes a column for fat in the dry matter (FDM) which is the target cheese fat content reported as a percentage of the target total solids content, where total solids is calculated as 100 minus the target moisture content. Because the principal nonfat component in cheese is casein, the target FDM value is useful to estimate the proportions of fat and protein required in the cheese milk. For example cheese makers generally consider a full fat cheese contains 50% FDM which corresponds to a protein fat ratio in the cheese milk of 0.94 - 0.96. By this criteria, both Cheddar and Feta are full fat cheese because they both contain about 50% FDM, although on a wet basis their respective fat contents are 31 and 22%.

P/F (ratio of protein to fat) is exactly what the name implies. Having no units, it is an index of the relative proportions of fat and protein in the milk. Please be clear that the P/F value indicates nothing about the absolute value of fat and protein. P/F ratio is generally lower in low fat milk and higher in high fat milk, so that Jersey milk, for example, has a less favourable P/F for cheese making than Holstein milk. This is partially offset by a higher casein number (casein as a percentage of total protein) in Jersey milk.

Standardization normally means adding skim milk or skim milk solids, or removing cream to increase the ratio of protein to fat (P/F). Several practical points are relevant.

Better process and composition control can be achieved by standardizing to fixed casein/fat ratios rather than protein/fat ratios. This requires accurate casein measurement which is still not feasible for most plants. See further discussion in Chapters 6 and 12.

Standardization usually requires the addition of protein or removal of fat. The former has the advantage that it is possible to produce cheese quantities beyond what's possible from the available fresh milk. This is significant in areas where fresh milk is in short supply or as in Canada, where milk purchases are limited by quotas. Several sources of milk proteins are available for cheese milk standardization.

(1) Skim milk powder is convenient for small or remote cheese plants. It can be used effectively with the following limitations:

(2) Skim milk and condensed milk are convenient sources because they can be handled and measured in liquid form. The only cautions are to limit heat treatment to minimum pasteurization requirements and limit nonfat milk solids to less than 11 kg/100 kg. Again, nonfat solids can be adjusted by adding water.

(3) Culture media contribute nonfat milk solids which must be accounted for in calculations for milk standardization. For example, the high heat treatment involved in bulk culture preparation ensures that most milk proteins (including whey proteins) present in the culture will be transferred to the cheese.

(4) Protein concentrates and isolates available to supplement cheese milk are numerous. A few are listed below. The feasibility of using one or more of these products, depends on, among other things, the type of cheese. For example, relative to most other varieties, high levels of whey proteins can be used in Feta cheese without compromising quality.

Most jurisdictions prohibit the use of non dairy fat in cheese. That leaves a number of choices:

In the absence of online systems equipped with customized algorithms, it is necessary to create spread sheets to calculate milk formulae and monitor yield parameters. The first step is to determine the optimum P/F, a process that always involves some experimentation. The estimates given in Table 6.1 can be used for a first approximation and then adjustments can be made on succeeding days based on the cheese analysis. This emphasizes the need for consistent and accurate records of milk and cheese composition and manufacturing parameters.

Detailed procedures, including calculations, for manual standardization are described in Chapter 6.

Automated composition control systems separate warm milk into cream and skim and then automatically and continuously recombine the two streams in the proportion required to obtain the desired P/F ratio. The standardized milk is tempered to the correct setting temperature and delivered directly to the setting vats. Two general types of control are possible.

Considering the limitations described above for protein and fat sources, it is possible to manufacture cheese from recombined milk.

Many people assume that all dairy products in Canada, including cheese, are made from pasteurized milk. Not so; several alternatives are possible as outlined below. Note, however, that the Food and Drugs Act and Regulations, recognizes only two types of cheese with respect to milk heat treatment, namely, fully pasteurized milk and raw milk. That is, if the milk is not fully pasteurized the resulting cheese is considered raw milk cheese.

The process of homogenization reduces milk fat globule sizes from 1 - 15 micrometer to less than 2 micrometer (a micometer is 0.000,0001 m). The natural membrane on the fat globule is replaced by milk proteins, mainly caseins. This results in increased interaction between fat globules and the casein particles in the rennet gel. For some cheese homogenization is desirable:

With respect to most firm to hard ripened cheese, many workers have observed that cheese made from homogenized milk are too tough and firm after pressing. However, there is evidence that homogenization for Cheddar cheese making has some advantages if medium pressure (6.9 MPa) is used and if only cream (35% fat) is homogenized and subsequently blended with unhomogenized skim milk (Nair et al. 2001, Int. Dairy Journal 10:647).

(1) Calcium Chloride is frequently added at a level of about 0.02% to aid coagulation and reduce amount of rennet required, especially if milk is set immediately after pasteurization. The role of calcium in milk coagulation will be discussed in Chapter 8.

(2) Nitrates (sodium or potassium nitrate) is added at levels of about 200 ppm to Edam, Gouda, Swiss to inhibit growth of gas forming Clostridium tyrobutyricum.

(3) Annatto cheese color is added to some cheese to standardize seasonal changes in color or to create orange cheese such as Cheddar and Cheshire.

Goats' and sheeps' milk are flat white in color because they lack -carotene. Cows' milk may be whitened to mimic goats' or sheeps' milk. Legal whitening agents include:

A wide range of products are available to accelerate cheese ripening or to develop a broader flavour profile. Relative to traditional cheese varieties, several factors suggest the need for ripening supplements:

Lipases (lipolytic enzymes) are traditionally added to cows' milk to produce cheese such as Feta, Romano, Kefalotyri, and Parmesan which are traditionally made from goats' or sheeps' milk. That's because goats' and sheep's milk, especially goats' milk, have more natural lipase than cows' milk. Commercial lipases are commonly extracted from kid goats.

Mixtures of enzymes from various sources added to the milk to accelerate ripening of aged cheese such as Cheddar. These cocktails include both lipases and proteases, with a predominance of proteases for Cheddar. Bacterial enzyme extracts from lactic acid bacteria have also been used. Accelerated ripening is further discussed in Chapter 10.

6. STANDARDIZATION OF MILK FOR CHEESE MAKING

Standardization refers to the practice of adjusting the composition of cheese milk to maximize economic return from the milk components while maintaining both cheese quality and cheese composition specifications. Composition specifications may be self imposed (eg., low fat cheese) or imposed by government standards of identity. In Canada, standards of identity are defined for 46 cheese varieties (Table 6.1). These standards only include limits for maximum moisture and minimum fat so they do little to standardize other cheese characteristics. For example, American Mozzarella is made by a different process and has different properties than Italian stretch Mozzarella, but by Canadian regulations American Mozzarella can be called Mozzarella provided it contains less than 52% moisture and more than 20% fat.

6.1 Important parameters of composition

Standardization of cheese milk normally requires increasing the proportion of protein relative to fat, which can be done by adding protein or taking away fat. The relative amount of protein and fat in milk is called the protein-fat ratio or P/F. The P/F is the principal factor which determines the amount of fat in the cheese relative to other milk solids in the cheese. Because it is easy to measure cheese fat and total solids, the proportion of fat in the cheese is reported as (1) fat on a wet basis; and (2) the ratio of cheese fat to cheese total solids. This ratio is called 'fat in the dry matter' or F/DM. The F/DM in cheese is determined mainly by P/F of the milk but the percent moisture is also important. Because cheese whey contains soluble solids, higher cheese moisture means that more soluble solids (mostly non-fat solids) are also retained in the cheese so that the ratio of F/DM decreases. The target value of F/DM in the cheese is used to determine the first approximation of the P/F required in the milk to give the desired fat content of the cheese. See Table 6.1.

There is a third ratio, namely, casein number (CN), which we will use in the standardization procedures given below, but which is important to understand. Total protein content of cows' milk is about 3.3Kg/hL of which about 2.6 /Kg/hL is casein. The remainder is whey protein (about .7 Kg/hL) including about .1 Kg/hL of some nitrogenous compounds which are not true protein and are referred to collectively as non-protein nitrogen (NPN). Casein is mostly recovered in cheese (i.e., transferred from milk to cheese during cheese manufacture). Whey proteins remain soluble in whey so that only small amounts are recovered depending on how much whey is retained in the cheese. Casein content is, therefore, most relevant to cheese yield, so when cheese makers standardize milk on the basis of protein content, they are using total protein as an index of casein content. Direct measurement of casein would be better because the proportion of casein in total protein varies with breed, season, region and other factors. However, wet chemical analysis of casein is not feasible for most plants and rapid instrumental methods are still under development.

The percentage proportion of casein in total protein is referred to as the casein number (CN).

6.2 Methods of Standardizing

There are three methods of standardizing milk, namely:

1. Addition of concentrated non-fat milk solids (i.e., skim milk powder or condensed skim).

2. Addition of skim milk.

3. Removal of cream.

These methods are based on the assumption that the milk has a high fat content relative to the protein content. This is normally the case, so that cows' milk usually has excess fat over that required to produce a legal cheese. The exceptions are high fat cheese such as cream cheese or double cream blue cheese.

It is not always economical to standardize milk. The cheese maker must compare the costs of standardizing with the extra yield of cheese or cream. Many cheese makers simplistically assume that all they have to do is standardize milk to meet the official composition standards. But the objective of standardization is to maximize the total return from all milk components while meeting regulations and without compromising quality. If the value of butter fat is low relative to protein, it is more economical to sell the fat as cheese rather than as cream provided that the extra fat can be retained in the cheese without compromising quality.

6.3 Units

Raw milk composition for payment purposes is reported in units of Kg of component per hL of milk at 4C. This is referred to as weight over volume (w/v) measurement. Measurement in units of w/v is dependent on milk density which in turn is affected by both composition and temperature. Weight over weight (w/w) measurement (eg., Kg component per 100 Kg of milk) results in a significantly smaller value because the density of milk is more than 1 Kg/L. Measurement by w/w has the advantages that: (1) most wet chemical reference analyses used to calibrate milk analysers report composition in units of w/w; (2) w/w values are independent of milk temperature. However, milk composition for payment purposes is reported in units of w/v because the volume of milk is easily measured with dip sticks or volumetric meters. Weight measurement would require installation of farm bulk tanks on expensive load cells. Volume rather than weight measurement of milk and other liquids is also more convenient in the plant.

In any case, the important point with respect to accurate standardization is to ensure that all measurements and calculations use the correct units. When component estimates are given as percentages, the basis of measurement must be stated as w/w percent (eg., Kg fat per 100 Kg milk) or w/v percent (eg., Kg fat/hL of milk). In this manual composition values given in percent always mean w/w. Cheese composition will always be stated in percent w/w (eg., 30% fat in Cheddar cheese means 300 g fat per Kg cheese). Similarly, 3.3 % fat in milk means 3.3 Kg fat per 100 Kg milk. If weight over volume units are used I will always state the specific units, eg., 3.3 Kg/hL. Because composition of producer milk is reported to processors in units of Kg/hL and because milk metering systems are volumetric, I will usually report milk composition in units of Kg/hL.

It is important to ensure that milk analysers are calibrated in the appropriate units and the correct units are subsequently used for milk standardization calculations and calibration of automated standardizing systems. Wet chemical analysis is normally done by weight, so reference results for milks used to calibrate milk analysers are normally reported in units of percent by weight and it is convenient to calibrate milk analysers in percent by weight (eg., Kg/100 Kg). If required, w/v values can be estimated using the following equation.

w/v=w/w x p^T where p^T is density at temperature T

Note, that the density must be known at the given temperature. For example, if the milk composition was given in units of w/w and you are metering milk into your cheese vat at 32C you need to know the density of the milk at 32C. For milk of average composition (4.0 % fat), the density can be estimated according to the following equation⁽¹⁾.

p^T = 1.0366 - .00035T where pT is density at temperature T

Density values for milk of average composition (4% fat) at some temperatures relevant to cheese manufacture are:

Temperature	4	10	20	32	37	40
Density	1.0352	1.0331	1.0296	1.0254	1.0237	1.0226

6.4 Calculations

The following steps are required to calculate the amount of powder or skim milk to be added, or the amount of cream to be removed. Suppose a cheese maker wishes to fill a 10,000 l (100 hl) setting vat for the manufacture of Cheddar cheese.

Step 1. Determine the protein and fat contents of the milk using an automatic milk analyzer. If a milk analyser is not available the protein content of pooled milk can be crudely estimated from the fat content using the following formula:

Kg/hL of protein = (0.4518 x Kg/hL of fat) + l.521

For the purpose of this example, assume the available milk contains 3.50 Kg/hL of fat and 3.l0 Kg/hL of protein.

Step 2. Determine the required fat, moisture and F/DM of Cheddar cheese. 'Dairy Products Regulations' of the Canada Agricultural Products Standards Act require Cheddar cheese to contain a minimum of 31% fat and a maximum of 39% moisture. Therefore,

F/DM = % fat/% dry matter = 30.0/(100.0 - 39.0) = 49.2%

Step 3. Determine the required P/F of the milk. The P/F required to yield F/DM = 50% as required for Cheddar cheese is about 0.96 (See Table 6.1).

Step 4. Calculate the amount of: skim milk powder to be added; or fat to be removed; or skim milk to be added.

Standardization by Adding skim milk powder

(i) Calculate the % protein required to give P/F = 0.96

The required level of protein = 0.96 x % fat = 0.96 x 3.50 = 3.36

(ii) The % protein to be added = 3.36 - 3.10 = 0.26 Kg/hL

(iii) Calculate the weight of protein which must be added per 100.00 hL of milk.

The required weight of protein = 0.26 Kg/hL x 100 hL = 26.0 Kg

(iv) Calculate the amount of powder which must be added assuming the skim milk powder (SMP) contains 35.0% protein. If possible the skim powder should be analyzed so the exact protein content is known. The supplier may be able to provide this information. Protein content can also be estimated using a milk analyzer to test the reconstituted skim milk.

Required amount of powder = 26.0 Kg/0.35 = 74.3 Kg

(v) Check calculations:

Weight of fat in milk:	3.50 Kg/hL x 100.00 hL = 350.0 Kg.
Weight of protein in milk	3.10 Kg/hL x 100.00 hL = 310.0 Kg
Weight of protein in SMP	0.35 Kg/Kg x 74.0 Kg = 26.0 Kg
Total Protein	310.0 Kg + 26.0 Kg = 336.0 Kg
P/F ratio of standardized milk	336.0 Kg/350.0 Kg = 0.96

Standardization by Removing fat

(i) Calculate the level of fat required to give P/F = 0.96.

The required level of fat = Kg/hL of protein/.96 = 3.10 Kg/hL/.96 = 3.23 Kg/hL

(ii) Use a Pearson's square to calculate the litres of cream that must be removed, assuming that the separator removes cream containing 30.00 kg/hl of fat.

Standardized Milk 3.23 Kg/hL		30.00 - 3.50 = 26.50 Parts Standardized Milk
	Unstandardized Milk 3.50 Kg/hl
Cream 30.00 Kg/hL		3.50 - 3.23 = 0.27 Parts Cream
	Total Parts	26.50 + 0.27 = 26.77

This means that the required proportions of cream and fresh milk are 0.27 and 26.77 parts, respectively, for a total of 0.27 + 26.50 = 26.77 parts. On a percent basis, the components are:

Standardized milk 100 x 26.50/26.77 = 98.99% w/v

30% cream 100 x 0.27/26.77 = 1.01% w/v

(iii) Calculate how much cream must be removed from 10,000 Kg of milk to provide standardized milk containing 3.23% fat.

Cream to be removed = 1.01% of 100 hL = 1.01 hL or 101 L.

(iv) Check calculations:

Weight of fat in milk:	3.50 Kg/hL x 100.00 hL = 350.0 Kg
Minus fat in cream	30.00 Kg/hL x 1.01 hL = 30.3 Kg
Weight of fat in standardized milk	350.0 Kg - 30.3 Kg = 319.7 Kg
Net volume of milk	100.00 hL - 1.01 hL = 98.99 hL
Weight of protein:	3.10 Kg/hL x 98.99 hL = 306.9 Kg
Protein/fat ratio	306.9/319.7 = 0.960

(v) Adjust the final weight for the quantity of cream removed. If you wish to fill the vat completely sum the vat capacity and the initial estimate of the cream to be removed and recalculate the required amount of cream.

Approximate total volume of fresh milk:	100.00 hL + 1.01 hL = 101.01 hL.
Weight of cream to be removed	1.00% of 101.01 hL = 1.01 hL
Final volume of standardized milk	100.00 hL - 1.01 hL = 99.99 hL

Standardization by Adding skim milk

The following calculation is based on the assumption that the protein content of the skim milk is the same as the protein content in the skim portion of the fresh milk to be standardized. This is exactly true only when the skim milk is derived from the same source as the fresh milk.

(i) Use a Pearson square to determine the relative proportions of skim milk and milk required to yield a fat content of 3.23% as calculated in Step B above.

Skim Milk 0.10 Kg/hL		3.5 - 3.23 = 0.27 Parts Skim Milk
	Standardized Milk 3.23 Kg/hl
Unstandardized Milk 3.50 Kg/hL		3.23 - 0.10 = 3.13 Parts Unstandardized Milk
	Total Parts	0.27 + 3.13 = 3.40

This means that 0.27 parts of skim are required for 3.13 parts of milk where the total mixture consists of 0.27 + 3.13 = 3.40 parts. On a percent basis, the mixture is:

Skim milk	100 x 0.27/3.4 = 7.9%
Unstandardized Milk	100 x 3.13/3.4 = 92.1%

(ii) Calculate the amount of skim and fresh milk required.

Weight unstandardized milk	92.1% of 100 hL = 92.10 hL.
Weight of O.1% skim milk	7.9% of 100 hL = 7.90 hL.

(iii) Check:

Weight of fat in unstandardized milk	3.50 Kg/hL x 92.10 hL = 322.4 Kg
Weight of fat in skim milk	0.10 Kg/hL x 7.90 hL = 0.80 Kg
Total fat	322.4 Kg + 0.8 Kg = 323.2 Kg
Weight of protein	3.10 Kg/hL x 100 hL = 310.0 Kg.
Protein/fat ratio	310.0/323.2 = 0.959

6.5 Addition Of Cream

The natural P/F of milk is higher in low fat milk. In practice, this means that when the milk fat is less than 3.0%, it may be necessary to add fat to obtain P/F = 0.96 and make a full fat cheese with F/DM = 50%. When the required F/DM is less than 50%, it is unlikely that fat would have to be added to the milk. The natural P/F is also high in the fall and early winter so fat may have to be added for full fat cheese at these times. Some cheese such as double cream Blue or double cream Havarti may also require addition of fat. Given the fat content of available cream, a Pearson's square can be used to calculate the amount of cream required in a similar manner to the examples given above.

6.6 General Guidelines for Standardization

Determine the present composition of your cheese

Determine the fat and protein content of milk accurately and daily.

Measure milk volume or weight accurately and keep accurate records.

If powder is being added, use only high quality, low temperature, antibiotic free powder of known protein content. Low temperature powder is required to ensure that excessive denaturation of whey proteins in the powder will not impair milk coagulation and/or cause texture defects in the cheese. To ensure low temperature powder, ask your supplier to certify a whey protein nitrogen index (WPI) greater than 6.0. Effects of heat treatment on coagulation and cheese quality are discussed in Chapter 8.

Weigh accurately the weight of powder or skim milk added or the weight of cream to be removed.

Determine the composition of the standardized cheese and if necessary adjust the proportions of fat and protein in the cheese milk on succeeding days.

If bulk starter is being added reduce the amount of protein added by the amount of protein in the culture.

The maximum recommended level of skim milk solids in cheese milk is 11%. Normal milk contains about 9% skim solids so the maximum level of additional skim solids is 2%. If standardization requires more it is recommended to standardize by removing fat or adding skim milk rather than by adding skim milk powder. Another alternative is to add some powder and then complete standardization by removing cream or adding skim milk.

Without sophisticated metering equipment it is difficult to obtain exact standardization. Provided you have a milk analyser, you can do a final check of milk composition after the milk is in the vat and then 'fine tune' the P/F ratio by adding skim solids or cream as required.

It is not possible to predict the exact composition of the finished cheese. However, when manufacturing conditions and milk composition are the same from day to day, it is possible to predict the composition of cheese with greater accuracy and the proportions of fat and protein in the cheese milk can then be 'fine-tuned' accordingly. It is, therefore, important to keep accurate records.

Be careful to use the correct units when calculating and weighing and metering. See Section 2.

Table 6.1. Some cheese varieties with some characteristics, composition and suggested ratio of protein/fat in standardized milk. Fat and moisture levels for most varieties correspond to definitions given in the Canadian regulations.

					Cheese Target Composition					Yield
	Texture	Washing	Salting	Rind	Fat	Moist	FDM	MNFS	Prot/Fat	% w/w
Alpina (Stella Alpina)	Semi-soft	Maybe warm	B or DS	Smear	27	46	50.0	63.0	0.90	11.5
Asiago	Firm to hard	None	B	Dry	30.0	40.0	50.0	57.1	0.93	10.1
Baby Edam	Firm	Warm wash	B	None	21.0	47.0	39.6	59.5	1.56	8.7
Baby Gouda	Firm	Warm wash	B	None	26.0	45.0	47.3	60.8	1.15	9.7
Blue	Soft to semi-soft	None	DC&DS	Smear or none	27.0	47.0	50.9	64.4	0.87	11.9
Bra	Firm to hard	None	B or DS	Dry	26.0	36.0	40.6	48.6	1.40	7.6
Brick	Semi-soft to firm	Usually warm	DC or DS	Smear or none	29.0	42.0	50.0	59.2	1.04	9.7
Brie	Soft	No	DS	Mould	23.0	54.0	50.0	70.1	0.86	14.0
Butterkase (Butter)	Semi-soft	Maybe warm	B	Smear	27.0	46.0	50.0	63.0	0.90	11.5
Caciocavallo	Firm to hard	Hot Stretch	B	Dry	24.0	45.0	43.6	59.2	1.17	9.8
Camembert	Soft	None	DS	Mould	22.0	56.0	50.0	71.8	0.86	14.7
Canadian Muenster	Semi-soft	Maybe warm	B or DS	Smear	27.0	46.0	50.0	63.0	0.90	11.5
Cheddar	Firm	None	DC	None	31.0	39.0	50.8	56.5	0.91	10.0
Cheshire	Firm	None	DC	None	30.0	44.0	53.6	62.9	0.79	11.9
Colby	Firm	Cold wash	DC	None	29.0	42.0	50.0	59.2	1.03	9.7
Coulommiers	Soft	None	DS	Mould	22.0	56.0	50.0	71.8	0.85	14.8
Danbo	Firm, small eyes	None	B,DS or DC	Smear or none	25.0	46.0	46.3	61.3	1.04	10.6
Edam	Firm	Warm wash	B	Dry or none	22.0	46.0	40.7	59.0	1.50	8.7
Elbo	Firm	None	DS or B	Dry or none	25.0	46.0	46.3	61.3	1.04	10.6
Emmentaler	Firm with eyes	None	B	Dry or none	27.0	40.0	45.0	54.8	1.13	9.1
Esrom	Semi-soft	Maybe warm	DS or B	Smear	23.0	50.0	46.0	64.9	1.04	11.5
Farmers	Firm	Cold wash	DC	None	27.0	44.0	48.2	60.3	1.11	9.7
Feta	Soft	None	DS	None	22.0	55.0	48.9	70.5	0.90	14.0
Fontina	Semi-soft to firm	Maybe warm	DS or B	Light smear	27.0	46.0	50.0	63.0	0.90	11.5
Fynbo	Firm,small eyes	?	B or DC	Dry	25.0	46.0	46.3	61.3	1.05	10.5
Gouda	Firm, small eyes	Yes	B	None	28.0	43.0	49.1	59.7	1.07	9.7
Guyere	Firm, eyes	No	B&DS	Light smear	28.0	38.0	45.2	52.8	1.14	8.7
Havarti	Semi-soft	Warm wash	B or DS	Smear or none	23.0	50.0	46.0	64.9	1.19	10.5
Jack	Semi-soft	Cold wash	DC	None	25.0	50.0	50.0	66.7	1.02	11.4
Kasseri	Firm to hard	Hot stretch	B	Dry	25.0	44.0	44.6	58.7	1.13	9.8
					Cheese Target Composition					Yield
	Texture	Washing	Salting	Rind	Fat	Moist	FDM	MNFS	Prot/Fat	%w/w
Limburger	Soft to semi-soft	Maybe warm	DS or B	Heavy smear	25.0	50.0	50.0	66.7	0.88	12.6
Maribo	Firm, small eyes	None	B or DS	Dry or none	26.0	43.0	45.6	58.1	1.09	9.8
Montasio	Firm	Usually warm	DS or B	Dry	28.0	40.0	46.7	55.6	1.19	8.7
Monterey	Firm	Cold wash	DC	None	28.0	44.0	50.0	61.1	1.04	10.0
Mozzarella (Italian)	Semi-soft to firm	Hot stretch	B	None	20.0	52.0	41.7	65.0	1.22	11.1
Mozzarella (Canadian)	Firm	Cold wash	DC	None	20.0	52.0	41.7	65.0	1.22	11.1
Muenster	Semi-soft	Maybe warm	B or DS	Light smear	25.0	50.0	50.0	66.7	0.88	12.6
Parmesam	Hard, grating	None	B&DS	Dry	22.0	32.0	32.4	41.0	2.02	6.1
Part Skim Mozz	Semi-soft to firm	Hot stretch	B	None	15.0	52.0	31.3	61.2	1.90	9.1
Part Skim Pizza	Semi-soft to firm	Hot stretch	B	None	15.0	48.0	28.8	56.5	2.20	7.9
Pizza	Semi-soft to firm	Hot stretch	B	None	20.0	48.0	38.5	60.0	1.42	9.5
Provolone	Firm	Hot stretch	B	None	24.0	45.0	43.6	59.2	1.17	9.8
Romano	Hard	None	B&DS	Dry or none	25.0	34.0	37.9	45.3	1.58	7.0
Samsoe	Firm, few eyes	None	B&DS	Dry or none	26.0	44.0	46.4	59.5	1.05	10.1
Tilsiter (Tilsit)	Firm	Usually warm	B or DS	Smear or none	25.0	45.0	45.5	60.0	1.08	10.2
Tybo	Firm, few eyes	None	B	Dry or none	25.0	46.0	46.3	61.3	1.04	10.6

CONSTANTS, ASSUMPTIONS AND LEGEND

1. All cheese composition and yield values are in units of percent by weight--including both cheese and standardized milk.

2. Estimation of yield and protein/fat ratios are based on principles and yield equations described by D.B. Emmons, C.A. Ernstrom, C. Lacrois and P. Verret. J. Dairy Science 73(1990):1365.

3. Whey solids in moisture was assumed to be 6.5% except for washed types when a value of 3.2% was used. For the purpose of yield calculations, pasta filata types (hot stretch) were considered to be unwashed. 75% of cheese moisture was considered available as a solvent for whey solids.

4. Conversion factors: Proportion of fat transferred from milk to cheese was 0.93

Amount of casein + minerals transferred to cheese was casein x 1.018

Casein number was 76.5

Washing: 'warm' means washing at temperatures near normal cooking temperatures (32-40C)

'cold' means wash water at temperature less than 20C is used to wash and cool the curd

'maybe warm' means that the cheese may or may not be washed with warm water

'hot stretch' means the cheese is heated and worked in hot water (70-80C) as in Pasta Filata types.

Salting: B = brine salted; DS = dry salted on cheese surface; DC = curd dry salted before hooping.

FDM = fat as percentage by weight of cheese solids; MNFS = moisture as percentage of non-fat substance in cheese.

Prot/Fat = ratio of protein to fat in standardized cheese milk.

	Cow	Dairy Sheep	Water Buffalo	Goat
Fat	3.9	7.2	7.4	4.5
Total Protein Casein Whey	3.3 2.6 0.7	4.6 3.9 0.7	3.8 3.2 0.6	3.2 2.6 0.6
Lactose	4.6	4.8	4.8	4.3
Ash	0.7	0.9	0.8	0.8
Total solids	12.5	17.5	16.83	12.8

Table 4.2 The principal caseins and some properties of importance to cheese making
Name	Symbol	Percent of casein	Properties
Alpha-S1 casein	_S1	33	Binds calcium strongly Sensitive to break down by rennet Resists the natural milk protease, plasmin
Alpha-S2 casein	_S2	11	Binds calcium strongly
Beta-casein		33	Partially soluble in cold milk Broken down by plasmin but not rennet
Kappa-casein		11	Stabilizes casein particles against coagulation Bonds with whey proteins during heating

Table 4.3 The principal whey proteins and some properties of importance to cheese making
Name	% of Whey Protein	Properties
Beta-lactoglobulin	40	Readily reacts with -casein at temperatures greater than 65C and interferes with rennet coagulation Principal component of ricotta cheese
Alpha-lactalbumin	15	Principal protein of breast milk Used to prepare infant formula
Immuno-globulins	6	Stabilizes casein particles against coagulation Bonds with whey proteins during heating
Other heat sensitive proteins	4.0	Mainly includes bovine serum albumin.
Heat stable proteins	14	Cannot be recovered by heat-acid precipitation as in ricotta cheese manufacture.
Non-protein nitrogen	21	Consists of amino acids, ammonia, urea and small peptides.

Table 4.4 Typical fat and protein contents (kg/100 kg) for the milk of several breeds of dairy cows (from various sources).
Breed	Fat	Protein	Protein/Fat Ratio
Jersey	5.4	3.8	0.70
Holstein	3.8	3.2	0.84
Guernsey	4.9	3.6	0.73
Ayrshire	4.0	3.3	0.83

Standardized milk	100 x 26.50/26.77 = 98.99% w/v
30% cream	100 x 0.27/26.77 = 1.01% w/v