Rice Grading and Brokens Separation

Grading exists to turn one mixed stream into several saleable ones. Rice is traded on grades defined largely by how much broken material a lot contains, and those grades carry substantially different prices. A stream that has not been graded cannot be sold as any of them: it must be sold at the value of its worst component. Separation by length is what converts a physical fact about the lot into a commercial position.

The principle rests on the one dimension breakage changes. Because breakage shortens a kernel without altering its width or thickness, length is the only property that distinguishes a broken from a whole grain — and length is precisely what a screen cannot measure, since a grain presents its cross-section to an aperture rather than its length. The indent cylinder measures length by fit: a pocket of a given depth accepts a kernel shorter than itself and rejects a longer one, and rotation converts that fit into a physical lift that carries the short kernels away from the stream. Changing the pocket size changes where the cut falls, which is how a mill produces several broken fractions rather than just two streams.

Grading yields head rice and one or more broken fractions. Head rice is not modelled here as a separate product: it is the whole-grain portion of milled rice, which is itself a traded commodity, and this reference keeps it within that commodity rather than creating a product entity for a fraction of it. The declared product output is broken rice, which is separated from the stream and sold into different markets from the grain it came from.

Broken rice is a genuine co-product, and calling it a defect fraction misreads its economics. It is food in its own right in many cuisines, where certain dishes are made with it by preference rather than by necessity. It is a raw material for rice flour, for brewing, and for starch. It is a feed ingredient. Its price sits below head rice — that is what makes breakage a loss — but it is a real market and not a disposal route, and in some regions demand for it is strong enough that mills produce it deliberately rather than merely accepting it.

Head rice

The whole-kernel fraction, and the basis on which milled rice grades are largely defined. Retained within the milled-rice commodity rather than modelled as a distinct product.

Broken rice

Kernels below the length cut, separated into one or more size fractions. A co-product with established food, rice-flour, brewing, and feed markets.

Brewers rice and fine brokens

The smallest fractions, cut off by finer pockets. Sold into brewing, flour, and feed uses. Fractions of the broken-rice stream rather than separate entities here.

Grading changes no kernel’s quality; it changes which lot a kernel ends up in. Its faults are therefore misclassification faults, and they run in both directions with asymmetric consequences. Brokens carried into the head-rice fraction put the grade at risk of failing its specification, which is a commercial and contractual problem. Whole kernels lifted out into the broken fraction are simply value sold at a discount — quieter, harder to notice, and easy to tolerate for far longer than it should be.

Because the separation depends on the geometry of the pocket, its accuracy degrades as that geometry wears, and it depends on the grain being what the pockets were chosen for. A cylinder indented for one variety’s kernel length does not sort another variety’s correctly, which is why grading is set against the rice being run and not once and for all.

• Whole kernels lifted into the broken fraction — value sold at a discount
• Brokens carried into head rice, putting the grade specification at risk
• Pocket wear degrading the sharpness of the length cut over time
• Cylinder indent size mismatched to the variety’s kernel length
• Additional breakage from handling and conveying between and after separations

Grading is among the least energy-demanding steps in a rice mill, since it only turns cylinders and moves grain rather than removing material. It generates no residual stream at all: every fraction it makes is sold, and the separation exists precisely so that each can be. In material terms nothing leaves the mill here that is not a product.