Sieve Analysis

A weighed working sample is placed on a sieve, or a nested stack of sieves of progressively different aperture sizes, and shaken — by hand for a prescribed time and pattern, or by a mechanical shaker set to a prescribed motion and duration. Material too large or the wrong shape to pass through a given aperture is retained on that sieve; everything else passes through, either to a finer sieve beneath it or into a collecting pan. Each fraction is weighed and reported as a percentage of the total sample weight.

This is why the applicable grade standard or official method specifies aperture size and shape, sieving time, motion, and sample mass in precise, prescriptive detail rather than leaving them to the operator's discretion: those choices are not incidental to the test, they are the test.

Sieving sorts material by physical dimension and shape alone. A weed seed, a small stone, and a broken kernel that happen to share similar dimensions all end up in the same retained or passed fraction, indistinguishable to the sieve itself. Turning a sieved fraction into a meaningful classification — how much of it is genuinely broken grain, how much is foreign material, how much is a particular size class of sound kernel — requires the fraction to be examined and, commonly, hand-sorted afterward. The sieve performs the physical separation; a person or a further inspection step performs the classification.

• Under-sieving leaves material unseparated that the standard intends to separate; over-sieving abrades sound kernels and creates new fine material, inflating the fine fraction beyond what was actually present in the original sample — which is why time and motion are prescribed rather than left to judgement.
• A worn, deformed, or blinded sieve — one whose apertures have plugged with fine or fibrous material — returns a plausible-looking result with no built-in indication that the apparatus itself is at fault.
• Sample moisture affects how cleanly material separates: damp grain and damp fines adhere to each other and to the screen, so the identical lot can sieve differently wet than dry.

None of these effects are visible in the resulting percentage figure itself; a sieve analysis result looks equally confident whether the procedure was followed exactly or not, which is why laboratories check sieve condition and follow the prescribed time and motion precisely rather than treating them as approximate.