Sieve Analysis and Gradation Test

One stack of sieves tells you what a soil or aggregate actually is: how much of it is gravel, sand and fines, and whether the sizes work together or fight.

the two boundary sieves: 4.75 mm splits gravel from sand, 75 μm splits sand from fines
StandardsASTM C136 for aggregate, D6913 for soil
Fines boundary75 μm, the No. 200 sieve
Sand and gravel boundary4.75 mm, the No. 4 sieve
Key diameters from the curveD10, D30 and D60, the sizes 10, 30 and 60 percent pass
Uniformity coefficientCu = D60 over D10
Curvature coefficientCc = D30 squared over D60 times D10
Well graded sandCu of 6 or more with Cc between 1 and 3
Well graded gravelCu of 4 or more with Cc between 1 and 3

How the test works

Wash the sample over the 75 μm sieve first so the fines that cling to coarser particles actually leave, dry it, then shake it through the nested stack and weigh what each sieve retains. Convert to cumulative percent passing and plot against size on a log scale. The shape of that curve is the material: a steep curve is uniform, single sized stuff, a long gentle curve is well graded, and a flat step means a missing size range, the signature of a gap graded material.

Where it goes wrong

Skipping the wash is the classic flattering error: unwashed gradations always read cleaner than the truth, and the missing fines show up later as pumping subgrade or dirty concrete aggregate. Overloaded sieves blind and stop passing what should pass, so split big samples. And remember what the test cannot say: gradation counts particles, not behaviour. The moment meaningful material passes the 200, the fines need Atterberg limits before anyone claims to have classified the soil.

Related Calculators and Reference