IS 2720:1985 Part 4 is the Indian Standard (BIS) for methods of test for soils - grain size analysis. This part of IS 2720 prescribes methods for the quantitative determination of the particle size distribution in soils. It details the mechanical sieving method for coarse-grained soils (larger than 75 microns) and sedimentation techniques (pipette and hydrometer methods) based on Stokes' Law for fine-grained soils.
Describes methods for determining the quantitative distribution of grain sizes in a soil sample by sieve analysis and hydrometer analysis.
Sieve sizes, sample masses, hydrometer procedure, dispersing agent dosage, sedimentation timings and IS soil-fraction classification.
| Reference | Value | Clause |
|---|---|---|
| Test | Grain-size analysis (sieve + sedimentation) | Cl. 1 |
| Sieve set — coarse-grained portion | 75, 40, 20, 10, 4.75 mm | Cl. 3 (Table 1) |
| Sieve set — fine-grained portion | 2 mm, 1 mm, 600, 425, 300, 212, 150, 75 µm | Cl. 3 (Table 1) |
| Sample mass — soil with ≤ 4.75 mm | ≥ 200 g | Cl. 4.1 |
| Sample mass — soil with up to 75 mm particles | Up to 60 kg (table-graded) | Cl. 4.1 |
| Wet sieving — washed through 75 µm | Yes (when fines > 5 %) | Cl. 4.2 |
| Hydrometer method — required when finer than 75 µm > 10 % | Yes (sedimentation) | Cl. 5 |
| Hydrometer — type | ASTM 152 H or IS-equivalent (1.000 – 1.030 sg) | Cl. 5.1 |
| Sedimentation cylinder | 1000 mL graduated, 1000 mm tall | Cl. 5.1 |
| Dispersing agent — sodium hexametaphosphate | 33 g + 7 g sodium carbonate per litre (33 mL of solution) | Cl. 5.2 |
| Soaking time before agitation | ≥ 16 hours | Cl. 5.3 |
| Hydrometer reading times (typical) | 0.5, 1, 2, 4, 8, 15, 30, 60, 120, 240, 1440 min | Cl. 5.4 |
| Particle-size classification — gravel | >2 mm (4.75 mm coarse boundary in IS soil class.) | Cl. 6 |
| Particle-size — sand | 0.075 – 4.75 mm | Cl. 6 / IS 1498 |
| Particle-size — silt | 0.002 – 0.075 mm | Cl. 6 / IS 1498 |
| Particle-size — clay | < 0.002 mm | Cl. 6 / IS 1498 |
| Stokes' law constant — used in calc | v = (γs−γw)·D²/18η | Cl. 5.5 |
| Result reporting — particle-size distribution curve | Semi-log plot (% passing vs log D) | Cl. 6.1 |
IS 2720 (Part 4) specifies the method for grain-size analysis of soils — the foundational geotechnical test that classifies a soil as gravel, sand, silt, or clay. Every soil report starts here.
Grain-size distribution drives: - Soil classification (IS 1498:1970, Indian Standard / Unified Soil Classification System) - Bearing capacity estimates (IS 6403:1981, IS 1888 plate load) - Compaction control (IS 2720 Part 7 / Part 8) - Permeability and seepage estimates - Filter design (Terzaghi filter rules) - Pavement subgrade design (IRC:37:2018)
Use IS 2720 Part 4 whenever you need to characterise a soil for foundation design, embankment construction, pavement subgrade, retaining wall backfill, or seepage analysis. It's run on samples from boreholes, trial pits, or borrow areas during the geotechnical investigation phase.
IS 2720 has 41 Parts in total — Part 4 is the gradation test. Parts on consistency limits, density, shear strength, consolidation, permeability, etc. are separate.
Method A — Wet sieving for coarse fraction (> 75 µm) 1. Air-dry sample, weigh 500 g (sand) to 5 kg (gravel) 2. Soak in water with sodium hexametaphosphate dispersant for 12 hours 3. Wash through 75 µm sieve, retain both wash-water (with fines) and oven-dry the > 75 µm fraction 4. Sieve the dry > 75 µm fraction through standard sieves: 80, 40, 20, 10, 4.75, 2.36, 1.18, 0.6, 0.3, 0.15, 0.075 mm 5. Record cumulative percentage passing each sieve
Method B — Sedimentation (hydrometer or pipette) for fines (< 75 µm) 1. Take 50 g of < 75 µm fraction (or whole soil if very fine) 2. Disperse in distilled water + sodium hexametaphosphate (40 g/L) overnight 3. Transfer to 1000 mL sedimentation cylinder; agitate end-over-end 4. Hydrometer method: lower the calibrated hydrometer (IS 3104) at 1, 2, 4, 8, 15, 30, 60, 240 min and 24 hr; read suspension density 5. Pipette method (more accurate): extract 25 mL at standard depth at the same time intervals; oven-dry, weigh 6. Use Stokes' law to convert settling velocity to particle diameter
Report: complete cumulative grain-size distribution curve, conventionally plotted on a semi-log graph (% passing on linear y-axis vs particle size on log x-axis).
Particle size definitions (IS 1498):
| Fraction | Size range | |---|---| | Boulder | > 300 mm | | Cobble | 80-300 mm | | Gravel | 4.75 mm - 80 mm | | Sand | 75 µm - 4.75 mm | | Silt | 2 µm - 75 µm | | Clay | < 2 µm |
Coefficients computed from grading curve: - D₁₀ = particle size at which 10 % is finer (effective size) - D₃₀ = at which 30 % is finer - D₆₀ = at which 60 % is finer - C_u = D₆₀ / D₁₀ (Coefficient of Uniformity) — well-graded sand: C_u > 6; well-graded gravel: C_u > 4 - C_c = D₃₀² / (D₁₀ × D₆₀) (Coefficient of Curvature) — well-graded if 1 ≤ C_c ≤ 3
Soil classification thumb-rules (IS 1498:1970): - > 50 % retained on 75 µm: coarse-grained (gravel or sand) - > 50 % passing 75 µm: fine-grained (silt or clay) - Within fine-grained, plasticity classification uses IS 2720 Part 5 (Liquid Limit + Plastic Limit)
Filter criteria (Terzaghi for granular filters): - D₁₅(filter) ≥ 4 × D₁₅(base) — permeability - D₁₅(filter) ≤ 4 × D₈₅(base) — piping prevention - C_u(filter) ≤ 20
1. Skipping the sodium hexametaphosphate dispersant on clayey soils. Without dispersant, clay particles flocculate and settle as larger 'pseudo-silt' — the fines content reads artificially low and you mis-classify the soil. 2. Wet-sieving losses on the 75 µm. If the wash water is not collected and the fines content not added back from sedimentation, the gradation curve breaks at 75 µm. Always reconcile dry mass against the original sample. 3. Hydrometer temperature correction. Stokes' law assumes 27 °C reference; if test is at 20 °C or 35 °C, viscosity and density differ. Apply temperature correction per IS 2720 Part 4 Annex. 4. Reading the hydrometer at the wrong meniscus. Read at the top of the meniscus (the rim) for soil suspensions; bottom for clear water. The standard is explicit; mix-ups cause ~3-5 % systematic error. 5. Insufficient sedimentation time for clays. Some montmorillonitic clays take 48 hr to reach 2 µm; the standard 24 hr reading is conservative. For sensitive applications (filter design, dam core), extend to 48 hr. 6. Reporting sand+silt+clay percentages but not the curve. Two soils with same fractions but different curve shape behave very differently. Always include the plot. 7. Using one borrow-area sample to characterise an embankment. Borrow areas have heterogeneity. Take ≥ 3 samples per visible material change and average; for highway works, IRC:36:2010 specifies sampling density.
Standard geotechnical investigation cascade for a building / infrastructure project:
1. Site reconnaissance — visual, existing borings, geological maps. 2. Boring + sampling (IS 1892:1979) — disturbed and undisturbed samples at each major stratum. 3. Index tests on every stratum: - Natural moisture content (Part 2) - Grain-size analysis (Part 4 — this code) - Atterberg limits (Part 5) - Specific gravity (Part 3) - Classification per IS 1498 4. Strength tests on selected strata: - UCS (Part 10) - Direct shear (Part 13) or triaxial (Part 11) - SPT N-value via IS 2131:1981 at depth intervals - CBR (Part 17) for pavement subgrade 5. Compressibility — consolidation (Part 15) for soft clays / silts. 6. In-situ tests — plate load (IS 1888), pile load (IS 2911 Part 4). 7. Synthesis report — foundation type recommendation, allowable bearing pressure, settlement estimate, fill / borrow material grading specifications.
Grain-size analysis is the cheapest test in the chain (₹500-1500 per sample) — never skip it, and run it on every stratum, not just the founding layer.