IS 2720:1983 Part 1 is the Indian Standard (BIS) for methods of test for soils: part 1 preparation of dry soil sample for various tests. This part of IS 2720 specifies the procedures for preparing dry soil samples from field-collected bulk samples. It covers the methods of drying, breaking clods, sample reduction via quartering or riffling, and sieving to specific fractions necessary for subsequent laboratory geotechnical tests.
Specifies methods for preparing representative dry soil samples suitable for various laboratory tests.
Key reference values — verify against the current code edition / project specification.
| Reference | Value | Clause |
|---|---|---|
| Subject | Preparation of dry soil samples for tests | Scope |
| Drying | Air-dry (or controlled) — avoid over-heating clays | Method |
| Quartering | Riffling/quartering for representative sub-sample | Method |
| Why | Sample prep error invalidates all downstream tests | Importance |
| Read with | IS 2720 other parts (the actual tests) | Cross-ref |
IS 2720 (Part 1):1983 specifies the Methods of Test for Soils — Preparation of Dry Soil Samples for Various Tests. It is the foundational sample-preparation methodology that every subsequent IS 2720 Part references for sample handling before specific tests are run.
Use it when: - Preparing soil samples for any laboratory test in the IS 2720 series (Parts 2-41+) - Auditing lab quality — sample-prep is the most common error source; verify lab procedures match Part 1 - Specifying borehole sampling protocols — site investigation specs reference Part 1 for sample handling - Reviewing geotechnical reports — sample-prep deviations affect every downstream test result
This code is methodological infrastructure — it doesn't give acceptance limits or design parameters, but every test that does (compaction Part 7/8, grain-size Part 4, Atterberg Part 5, shear Parts 11/13, etc.) assumes the sample was prepared per Part 1. Sample-prep errors propagate through every test.
The 33-part IS 2720 family: - Part 1: Sample preparation (this code) - Part 2: Water content / moisture content - Part 3: Specific gravity - Part 4: Grain size analysis (sieve + hydrometer) - Part 5: Liquid limit and plastic limit - Part 6: Shrinkage limit - Part 7-8: Compaction (Standard / Modified Proctor) - Part 11-13: Shear strength (triaxial / direct shear) - Part 16: California Bearing Ratio (CBR) - Part 17: Consolidation test - Part 28-29: Field density (sand replacement / core cutter) - Plus many more specialised tests
Two sample types covered (Clause 3):
Disturbed samples: where structural integrity of the soil isn't critical (used for classification, gradation, Atterberg limits, compaction tests). Sampled from boreholes, test pits, or remoulded in lab.
Undisturbed samples: where structural integrity matters (used for shear strength, consolidation, permeability, soil structure investigation). Sampled with thin-walled (Shelby) tubes or piston samplers to minimize disturbance.
Field handling: samples must be: - Labelled at source with depth, location, date, sampler ID, sample number - Sealed against moisture loss — disturbed samples in plastic bags + cloth bags; undisturbed in waxed tubes - Transported within 7 days to the lab in a temperature-controlled environment (avoid > 40°C exposure) - Stored at 27 ± 2°C, 90%+ RH in the lab humidity room until testing
Sample-prep steps in the lab (Clause 4):
1. Air drying (Clause 4.2): spread sample in shallow trays on absorbent paper; air-dry in temperature-controlled room until friable. AVOID oven-drying for plasticity tests (Atterberg limits) — oven heat alters clay-mineral structure irreversibly, giving falsely low LL/PL values.
2. Mechanical breaking (Clause 4.3): break aggregates / clods by rubber-padded hammer or roller. Don't crush individual mineral grains — that changes the gradation.
3. Riffle splitting / quartering (Clause 4.4): reduce the bulk sample to required size for each test without bias. Riffle splitter is preferred over coning-and-quartering — gives more representative sub-samples.
4. Pre-test conditioning (per individual Parts): - For LL/PL (Part 5): mix sample with distilled water to a paste consistency; rest 24 hours covered to allow moisture distribution - For sieve analysis (Part 4): wash sample over 75-micron sieve to separate fines from gravel/sand - For compaction tests (Parts 7/8): pre-condition at target moisture content; rest 16-24 hours sealed - For shear tests (Parts 11-13): cut from undisturbed tube, trim to size, mount in apparatus
1. Oven-drying samples meant for plasticity tests — Part 1 specifically warns: oven-drying changes the Atterberg limits of clayey soils. Air-dry only. Many labs short-cut this; reported LL/PL values from oven-dried samples are 10-30% lower than actual, leading to misclassification.
2. Inadequate sample size — Part 1 specifies minimum sample masses for each test (Table 1): - Sieve analysis (gravel): 5 kg minimum - Sieve analysis (sand): 500 g - Hydrometer: 50-100 g - LL/PL: 200 g - Compaction: 16-20 kg - Triaxial / CBR: 30-50 kg Sites sometimes send 'a small bag' to the lab thinking it's adequate; lab can't do all the tests; lab improvises or sends back for more samples (delay).
3. Cross-contamination — using the same tools (mortar, splitter, sieves) on multiple samples without cleaning between samples carries fines from one sample to the next. Clean equipment between every sample.
4. Loss of moisture during transport — disturbed bag samples kept in sunlit truck-bed dry out before they reach the lab. Use sealed bags; transport in shaded / temperature-controlled vehicle.
5. Undisturbed sample disturbed — Shelby tubes dropped, vibrated, or stored upside-down lose their structural integrity. Handle with care; mark 'TOP' and 'BOTTOM'; wax-seal both ends; store vertically.
6. Identification labels lost — sample bags with internal labels get water-damaged; external pen-marks fade. Use waterproof labels; double-label (internal + external); maintain field log book.
7. Skipping the 24-hour rest for Atterberg samples — clay particles need time to absorb water uniformly. Testing within 30 minutes of mixing gives erratic LL/PL; 24-hour rest gives consistent results.
8. Riffle splitter not balanced — splitter must give 50/50 division within ± 5%; bias in the splitter introduces sampling error. Verify periodically.
9. Not following ANY sample-prep protocol — many small-town labs improvise. Two labs testing the same field sample report different results because they prepared the sample differently. Standardisation to IS 2720 Part 1 is the only way to ensure reproducibility.
The cost of soil sample preparation done badly: - CBR test on poorly-prepared sample: CBR varies by 50-100% from prep variations. A specification of CBR ≥ 4% can pass or fail by the prep alone. - Atterberg limits on oven-dried clay: LL drops from 60 to 45; sample classification changes from CH (fat clay, expansive) to CL (lean clay, ordinary). Foundation design implications: under-reamed piles vs ordinary footings — 5× cost difference. - Sieve analysis without washing: fines underestimated; soil misclassified; subgrade design wrong; pavement fails. - Compaction test on segregated sample: MDD over- or under-estimated; field compaction target wrong; long-term settlement or shear failure.
For project geotechnical reports: always specify in the tender / scope that lab testing must follow IS 2720 Part 1 protocols. For critical projects, audit lab sample-prep practices (some labs have improved standardised SOPs; many don't).
For consulting geotechnical engineers: when reviewing third-party reports, the sample-prep section often goes unread but is the source of most test-result variability. Ask: what was the air-dry vs oven-dry protocol? Was the riffle splitter calibrated? Was undisturbed sample stored correctly? These questions catch the bad reports.
IS 2720 Part 1:1983 is 42 years old but methodologically sound. The principles of soil sample preparation are timeless; revisions would be incremental at best. No public sign of an update from BIS CED 43 sectional committee.
Indian geotechnical lab reality: - NABL-accredited labs (CSIR-CRRI, IIT geotech labs, specialist consulting firms): consistently follow IS 2720 Part 1 protocols. Test results reproducible within ± 5-10%. - University / college labs: variable. Equipment may be old; protocols may have drifted. Acceptable for routine educational work; not for high-stakes project decisions. - Commercial QC labs at construction sites: vary wildly. Top end (Geo-Foundations, Tandon Consultants, Tilak) maintains discipline. Bottom end produces fictional results.
For project geotechnical investigations: - Specify NABL-accredited lab in the tender - Audit lab credentials and SOP documentation before award - For high-stakes (foundation design > ₹50 lakh cost, infrastructure projects): consider duplicate testing at two labs and compare results — divergence > 15% indicates sample-prep or technique issues
For consulting reviews: - Read the sample-prep section of any geotechnical report carefully - Cross-reference reported LL/PL against the soil type (CH/CL/MH classification) — inconsistencies indicate oven-drying issue - Check that 'undisturbed' samples were actually undisturbed (Shelby tube handling photos, in-tube preservation method)
Bottom line: Part 1 is unglamorous infrastructure code that determines the validity of every downstream test. Skipping or shortcutting it invalidates the entire geotechnical investigation. Insist on its proper application.