Buy a full case and the price per unit drops automaticallyRequest a sample
AccountCart0
All ContainersBottlesJarsCaps & ClosuresPumps & SprayersTubs & PailsJugsDroppersTinsTubesAccessories
Industries
Learn
Blending

Scaling a formula without drifting

Updated

On this page5

The first time you double a batch it usually works. The third time you double it, the blend is not the same, and the formula on the card has not changed.

Percentages, not weights

Write every formula as percentages by weight of the total, not as grams. A percentage scales arithmetically and survives a change in batch size, a change in packaging and a change in the person making it.

Keep the batch weight as a separate line so that the actual quantities are derived rather than remembered. Somebody will otherwise scale five of six components correctly.

Mixing does not scale linearly

This is the part that catches people. The optimum mixing time and speed have to be evaluated for the blend, and they do not scale with batch size. A mixer that reaches uniformity in four minutes at two kilograms does not reach it in eight at four.

It may need less time proportionally, because fill level changes how material tumbles. It may need more. The only way to know is to sample the larger batch and look.

And over-mixing is a real failure rather than a wasted minute: excess energy breaks leaf down, generates fines, and widens the distribution that causes segregation later.

What changes when the batch grows

Fill level in the mixer, which changes the tumbling action. Time to discharge, during which segregation is already starting. Depth of the holding container, which increases percolation. And your own handling, because a larger batch gets moved and scooped more.

None of those are in the formula and all of them affect the product.

Re-qualifying at the new size

Treat a scale-up as a new product until proven otherwise. Draw samples from top, middle and bottom of the finished batch and compare them, taste against a retained sample of the previous size, and only then release it.

Keep a retained sample of every batch, sealed and cold. It costs a jar and it is the only way to answer the question of whether the blend has drifted, which will otherwise be settled by whoever remembers it most confidently.

References

Barbosa-Cánovas et al., Dry mixing of food powders: effect of water content and composition on mixture quality, Journal of Food Engineering; segregation-control literature for dry blends.
Written from published specification and peer-reviewed work, not from a laboratory of our own. Your supplier's certificate of analysis, your own moisture readings and your state's rules override anything general said here.