(1) All soil and rock samples should be carefully classified, noting particularly those characteristics that have a bearing on the perviousness and stratification of the formation. Soil samples should be classified in accordance with the Unified Soil Classification System described in MIL-STD-619B. Particular attention should be given to the existence and amount of fines (material passing the No. 200 sieve) in sand samples, as such have a pronounced effect on the permeability of the sand. Sieve analyses should be made on representative samples of the aquifer sands to determine their gradation and effective grain size D10. The D10 size may be used to estimate the coefficient of permeabililty к. The gradation is required to design filters for wells, wellpoints, or permanent drainage systems to be installed in the formation. Correlations between к and D10 are presented in paragraph 3-4.
(2) Laboratory tests depicted in figure 3-2 can be used to determine the approximate coefficient of permeability of a soil or rock sample; however, permeabilities obtained from such tests may have little relation to field permeability even though conducted under controlled conditions. When samples of sand are distributed and repacked, the porosity and orientation of the grains are significantly changed, with resulting modification of the permeability. Also, any air en trapped in the sand sample during testing will significantly reduce its permeability. Laboratory tests on
samples of sand that have been segregated or contaminated with drilling mud during sampling operations do not give reliable results. In addition, the permeability of remolded samples of sand is usually considerably less than the horizontal permeability kh of a formation, which is generally the more significant k factor pertaining to seepage flow to a drainage system.
(а) Saturating the sample and allowing it to drain. Care must be taken to assure that capillary stresses on the surface of the sample do not cause an incorrect conclusion regarding the drainage.
(б) Estimating Sy from the soil type and D10 size of the soil and empirical correlations based on field and laboratory tests. The specific yield can be computed from a drainage test as follows: (3-і) where
Vy = volume of water drained from sample V = gross volume of sample The specific yield can be estimated from the soil type
(or D10) and the relation given in figure 3-3 or table 3-2.