IS 2720:1987 Part 18 is the Indian Standard (BIS) for methods of test for soils - part 18: determination of compaction properties (modified proctor). This test method covers the determination of the relationship between the moisture content and dry density of soils using heavy compaction (Modified Proctor test). It is essential for determining the Maximum Dry Density (MDD) and Optimum Moisture Content (OMC) for earthworks and pavement subgrades.
Specifies the modified Proctor test for determining the moisture-density relationship of soils for higher compaction energy.
Key apparatus dimensions, compaction effort parameters, and procedural limits for the Modified Proctor soil compaction test.
| Reference | Value | Clause |
|---|---|---|
| Rammer Mass— Value as per Amd. No. 1 (Aug 1987). Original code stated 4.9 kg. | 4.89 kg | Cl. 2.2 |
| Rammer Height of Drop | 450 mm | Cl. 2.2 |
| Rammer Face Diameter | 50 mm | Cl. 2.2 |
| Number of Compaction Layers | 5 | Cl. 4.1 |
| Mould Capacity (Small)— For soils where <25% by mass is retained on the 20 mm IS Sieve. | 1000 cm³ | Cl. 2.1.1 |
| Mould Internal Diameter (Small) | 100 mm | Cl. 2.1.1 |
| Mould Internal Height (Small) | 127.3 mm | Cl. 2.1.1 |
| Blows per Layer (Small Mould) | 25 | Cl. 4.1 |
| Soil Fraction for Small Mould | Passing 20 mm IS Sieve | Cl. 3.1 |
| Mould Capacity (Large)— For soils where >25% by mass is retained on the 20 mm IS Sieve. | 2250 cm³ | Cl. 2.1.2 |
| Mould Internal Diameter (Large) | 150 mm | Cl. 2.1.2 |
| Mould Internal Height (Large) | 127.3 mm | Cl. 2.1.2 |
| Blows per Layer (Large Mould) | 56 | Cl. 4.1 |
| Soil Fraction for Large Mould | Passing 40 mm IS Sieve | Cl. 3.1 |
| Detachable Collar Height | 60 mm | Cl. 2.1.1 & 2.1.2 |
| Oven Drying Temperature— For determining water content. | 105 to 110 °C | Cl. 2.6 |
| Balance Accuracy (Soil Mass)— For 1000 cm³ mould. | 1 g | Cl. 2.5 |
| Balance Accuracy (Soil Mass)— For 2250 cm³ mould. | 5 g | Cl. 2.5 |
| Balance Accuracy (Water Content) | 0.01 g | Cl. 2.5 |
| Formula for Bulk Density (ρ)— M = Mass of compacted soil; V = Volume of mould. | M / V | Cl. 5.1 |
| Formula for Dry Density (ρd)— ρ = Bulk density; w = water content as a decimal. | ρ / (1 + w) | Cl. 5.2 |
IS 2720 (Part 18):1987 specifies the Method of Test for Soils — Determination of Field Density by Sand Replacement Method using a small pouring cylinder, OR by the rubber balloon method, OR by water displacement — in-situ density tests for soils that cannot be conveniently undisturbed-sampled.
Use it when: - Quality control of earthwork — embankment fill, road subgrade, pipe trench backfill, building plinth fill — verify that field compaction achieved the specified % of MDD - Acceptance testing of compaction — typically 95% of Standard Proctor MDD (per IS 2720 Part 7) or 95-98% of Modified Proctor MDD for roads / dams - Investigating settlement issues — measure in-situ density of suspect zones; compare with adjacent good areas - Auditing third-party compaction work — independent verification
This is a field test, performed at the construction site. Companion codes: - IS 2720 Part 28:1974 — Sand replacement method using a large pouring cylinder (for coarser materials) - IS 2720 Part 29:1975 — Core cutter method (for cohesive soils only) - IS 2720 Part 32:1971 — Water displacement method (for rocky / stony soils) - IS 11315:1991 — Nuclear gauge method (faster alternative for routine roadworks QC) - IS 2720 Part 7:1980 — Standard Proctor (gives the MDD reference) - IS 2720 Part 8:1983 — Modified Proctor (alternative MDD reference for heavier compaction)
Method A — Sand Replacement with Small Pouring Cylinder (Clause 3):
*Apparatus*: small pouring cylinder (volume ~ 1.0 dm³), calibrated for clean uniform sand of known density (e.g., Indian standard sand of density ~ 1.55 g/cm³, IS 650).
*Procedure*: 1. Excavate a hole in the test surface — typically 100 mm diameter × 150 mm deep (for fine-grained soils) 2. Collect all excavated soil; weigh; calculate moisture content per IS 2720 Part 2 3. Fill the hole with calibrated sand from the pouring cylinder (sand level matches surrounding surface) 4. Weigh remaining sand in cylinder; calculate sand used = (sand to fill hole) 5. Compute hole volume = (sand used) / (sand density) 6. Compute in-situ wet density = (wet soil weight) / (hole volume) 7. Compute in-situ dry density = (wet density) / (1 + water content) 8. Compute % compaction = (in-situ dry density / MDD) × 100
*Limitations*: works best for fine-grained / sandy soils; not suitable for hard rock or very stony soils where sand can't flow into voids.
Method B — Rubber Balloon Method (Clause 4):
*Apparatus*: balloon-density apparatus with rubber bladder, water reservoir, and measuring scale.
*Procedure*: 1. Excavate hole as above 2. Place balloon apparatus over hole; inflate balloon with water; rubber balloon conforms to hole shape 3. Water displacement gives hole volume directly 4. Compute density as in Method A
*Advantages*: less operator-skill-dependent than sand replacement; faster. *Limitations*: equipment more expensive; not widely available in Indian site labs.
Method C — Water Displacement (rare; for very rocky soils): detailed in IS 2720 Part 32; uses heavy-duty rubber bladder.
Sample size requirements (Clause 5): - Wet weight ≥ 500 g for fine-grained soils - Wet weight ≥ 1500 g for granular soils - Hole depth ≥ 100 mm to give representative sample
Test frequency (per IRC SP 19 + IS 2720 Part 18 + project specifications): - Highway embankment compaction: 1 test per 500 m² of compacted layer; more frequent in critical areas (bridge approaches, retaining wall backfills) - Building plinth fill: 1 test per 100 m² per layer - Pipe trench backfill: 1 test per 50 m of trench per layer - Earthwork inspection at acceptance: 1 test per 500 m² minimum; more for safety-critical structures
Quality control: - Calibrate sand density with known reference volume monthly - Cross-check 1 in 20 tests by alternative method (rubber balloon, nuclear gauge) - Investigate any % compaction < specified value: re-test adjacent areas; if persistent, the compaction lift requires re-rolling
Typical acceptance criteria: - Building / general fill: ≥ 90% MDD (Standard Proctor) - Highway / road embankment: ≥ 95% MDD (Modified Proctor, per IRC 37 / MoRTH) - Earthen dam core: ≥ 98% MDD (Standard Proctor, with moisture control) - Subgrade for flexible pavement: ≥ 95% MDD (Modified Proctor) - Foundation grade fill (under shallow footings): ≥ 90% MDD
1. Wrong sand for sand replacement — calibration sand density varies (typical 1.40-1.60 g/cm³). Use only IS 650 standard sand of certified density; recalibrate periodically.
2. Hole edges not vertical — sloping hole walls cause sand to slump excessively, over-estimating hole volume and under-estimating in-situ density. Cut hole edges vertical with a knife; clean loose material before pouring sand.
3. Inadequate compaction of excavated surrounding material — when the hole is dug, vibration can settle surrounding ground. Use minimum-disturbance digging methods (gentle scraping; avoid impact tools).
4. Moisture content sample taken from wrong source — moisture content must represent the excavated material, not surface soil. Take moisture sample from the hole, AFTER excavation, before any drying / handling.
5. Sand pouring not at constant rate — too fast pour creates voids in the sand fill; too slow allows the sand to compact under its own weight. Use the pouring cylinder's calibrated drop technique consistently.
6. Single test for batch acceptance — one test gives a point reading; doesn't represent area-wide compaction. Test frequency per specification; investigate any low result with re-testing.
7. Field test not at MDD-test moisture state — if field moisture is significantly different from MDD/OMC reference, the apparent % compaction shifts. Sometimes lab needs to re-test MDD at field moisture for proper comparison.
8. No correction for oversize particles — if soil has aggregate > 19 mm (the sieve cut-off for Proctor MDD), the field test averages stone + matrix. Field density appears low not because matrix isn't compacted but because stone displaces matrix. Use AASHTO T-224 oversize correction or specify aggregate-free MDD reference.
9. Nuclear gauge as substitute without calibration — IS 11315 nuclear gauge is fast but requires field calibration against sand replacement results periodically. Don't substitute nuclear-only for major project acceptance; use sand replacement as referee.
10. Test not performed promptly after compaction — if test is delayed, soil dries or rebounds; density changes. Test within 24 hours of compaction, ideally same day.
IS 2720 Part 18:1987 is 38 years old but functionally adequate. The methods covered are mature international practice. No major revision in sight.
Indian site testing reality: - Highway and infrastructure projects: routine sand-replacement + nuclear-gauge testing at frequencies specified by IRC / MoRTH. NABL-accredited labs maintain quality. - Building / urban projects: variable. Many smaller projects don't conduct routine compaction tests; rely on "contractor's word" or limited owner audit. Recommendation: insist on independent third-party testing for any project > ₹50 lakh. - Government projects with strict QA: mandatory testing per specification; tests conducted by either project lab or NABL-accredited contractor labs.
Cost reality: - Sand replacement test: ₹200-500 per test (manual + small equipment) - Nuclear gauge: ₹50,000-2 lakh equipment; ₹50-150 per test (rapid) - Lab-side moisture and Proctor reference: ₹500-1500 (one-time per soil source)
For large infrastructure projects, nuclear gauge testing dominates — hundreds of tests per day on highway compaction work. Sand replacement is the referee test when results are disputed.
For specification documents: - 'Compaction shall be verified by IS 2720 Part 18 sand replacement OR IS 11315 nuclear gauge, with not less than 1 test per 500 m² per layer' - 'Minimum % compaction shall be ≥ X% of Standard/Modified Proctor MDD' - 'In case of dispute, sand replacement test per IS 2720 Part 28 (large cylinder, more accurate) shall be the referee'
Quality assurance recommendations: - Train field staff on proper testing technique (hole excavation, moisture sampling, sand pouring) - Cross-check 1 in 20 tests by alternative method - Investigate failing tests promptly; track failure rate (should be < 5% for well-supervised work) - Document test results with photos + GPS coordinates for traceability