Standardisation means adjusting milk to a target composition. The University of Guelph's cheese making text puts it plainly: it normally means adding skim milk or skim milk solids, or removing cream. The arithmetic that tells you how much is almost always a Pearson square.

It is a simple method, and it is worth understanding properly rather than using a template, because its one limitation is the source of a surprising number of off-specification batches.

How the square is set up

  1. Write the target value in the centre of a square.
  2. Write the composition of the two components you are blending at the two left-hand corners.
  3. Subtract across each diagonal, ignoring whether the result is positive or negative, and write the result at the opposite right-hand corner.
  4. Each right-hand figure is the number of parts of the component diagonally opposite it.

The target must lie between the two components. You cannot blend two milks to reach a fat content higher than both of them.

ILLUSTRATIVE VALUES: BLENDING TO 3.0% FAT Milk 4.5% fat Skim milk 0.1% fat Target 3.0% 3.0 minus 0.1 2.9 parts milk 4.5 minus 3.0 1.5 parts skim
Subtract across each diagonal and ignore the sign. The results are the parts of each component, which here total 4.4.

A worked example

The figures below are illustrative, chosen to be easy to follow. Use your own tested values in practice.

Suppose you have milk at 4.5% fat and skim milk at 0.1% fat, and you need 1,000 kg at 3.0% fat.

  • Milk: 3.0 minus 0.1 gives 2.9 parts
  • Skim milk: 4.5 minus 3.0 gives 1.5 parts
  • Total: 4.4 parts

Milk is 2.9 of 4.4 parts, or 65.9% of the blend. Skim milk is 1.5 of 4.4 parts, or 34.1%. For a 1,000 kg batch that is 659.1 kg of milk and 340.9 kg of skim milk.

Always check it by mass balance

The fat in 659.1 kg of milk at 4.5% is 29.66 kg. The fat in 340.9 kg of skim milk at 0.1% is 0.34 kg. Together that is 30.00 kg of fat in 1,000 kg, which is exactly 3.0%. If the check does not close, something in the setup is wrong, and it is far cheaper to find that on paper than in a tank.

The limitation that catches people

A Pearson square handles one component at a time. Blend two streams to hit a fat target and the square will land fat exactly, but the solids-not-fat of the result is simply whatever the two streams happen to produce.

Using the same illustrative streams, if the milk carried 8.7% SNF and the skim milk 9.0%, the blend would land at about 8.8% SNF. That may be fine. It may also be below what the product requires, and the square gave no warning either way.

Two targets need three components. Mixing two streams gives you one degree of freedom, enough to hit one target exactly. To land both fat and SNF, you need a third component, such as skim milk powder, and you solve the mass balances for total weight, fat and SNF together rather than using the square.

Solving for fat and SNF together

With three components of masses x, y and z, and a batch of mass M, three equations must hold at once:

  • Total mass: x + y + z = M
  • Fat: x times its fat % + y times its fat % + z times its fat % = M times the target fat %
  • SNF: the same expression, using each component's SNF % and the SNF target

Three equations and three unknowns have a single solution, provided the targets are achievable from the components chosen. If the solution demands a negative quantity of anything, the targets cannot be reached with those streams.

A note on what the target should be

For a named milk product, the minimum composition is set by the applicable food standards regulation for that product category, and those figures are periodically revised. Confirm the current requirement before standardising to it, rather than relying on a figure carried over from an older specification.