Codes referenced
The single IS code that names the aggregate properties a concrete producer or QA team must check.
Sampling, riffle reduction, sieve analysis, flakiness and elongation indices.
Silt content by decantation; clay lumps, coal/lignite, mica limits.
Pycnometer method for fine aggregate; wire-basket method for coarse.
Aggregate Impact Value and 10 percent fines value are the two limits IS 383 references.
Cl 5.3 names IS 383 as the aggregate source. IS 10262 reads aggregate properties (sieve, SG, absorption) for trial-mix design.
Why aggregate quality controls concrete quality
Aggregates occupy 70–80% of concrete volume. The cement paste is only the binder that welds the aggregate skeleton together; the aggregate is the load-carrying phase and the durability phase. A weak, dirty, badly-graded aggregate cannot be rescued by more cement — it can only be made more expensive by adding cement.
Three aggregate properties drive most field failure modes:
- · Grading — controls workability, water demand, and finish. Out-of-spec fines means more water, more shrinkage, lower strength.
- · Silt / deleterious content — coats aggregate surfaces and breaks the paste-aggregate bond. The classic cause of low cube strength with no visible problem in the mix.
- · Mechanical strength (AIV / 10% FACT) — controls the aggregate's resistance to impact during mixing, transport, placement, and (later) traffic loading.
IS 383 ties all three to numbers you can test. IS 2386 (Parts 1-4) gives you the test methods. IS 456 Cl 5.3 says "aggregates shall comply with IS 383." That is the audit chain — cite all three.
Fine aggregate — grading zones, silt, deleterious
Fine aggregate under IS 383 is material mostly passing 4.75 mm IS sieve, with a grading envelope that places it in one of four zones. The zones differ in fineness — Zone I is the coarsest, Zone IV the finest. IS 10262 and IS 456 do not mandate a zone, but the chosen zone shifts water demand and admixture dosage.
| IS sieve | Zone I (%) | Zone II (%) | Zone III (%) | Zone IV (%) |
|---|---|---|---|---|
| 10 mm | 100 | 100 | 100 | 100 |
| 4.75 mm | 90–100 | 90–100 | 90–100 | 95–100 |
| 2.36 mm | 60–95 | 75–100 | 85–100 | 95–100 |
| 1.18 mm | 30–70 | 55–90 | 75–100 | 90–100 |
| 600 µm | 15–34 | 35–59 | 60–79 | 80–100 |
| 300 µm | 5–20 | 8–30 | 12–40 | 15–50 |
| 150 µm | 0–10 | 0–10 | 0–10 | 0–15 |
Reference: IS 383:2016 Table 4 (graded fine aggregate — cumulative % passing by weight). All values are inclusive lower-upper limits.
Practical reading. Zone II covers most Indian river sands and many M-sands. Zone III is common for finer M-sand or crusher dust blends. Zone IV (finest) usually needs a higher paste volume and is rare as a single-source fine aggregate.
Silt, clay, deleterious
| Property | Method | Limit (IS 383:2016) |
|---|---|---|
| Silt / clay content (by decantation) | IS 2386 Pt 2 | ≤ 3% for M25 & below · ≤ 2% for above M25 · ≤ 1% for marine / aggressive exposure |
| Material finer than 75 µm | IS 2386 Pt 1 | As per grading table — usually ≤ 10% for Zone I-III |
| Clay lumps | IS 2386 Pt 2 | ≤ 1% by weight |
| Coal & lignite | IS 2386 Pt 2 | ≤ 1% by weight |
| Mica | IS 2386 Pt 2 | Desirable ≤ 1%; total lightweight ≤ 1% |
| Specific gravity | IS 2386 Pt 3 | Typically 2.6–2.7 — used in mix design calculation |
| Water absorption | IS 2386 Pt 3 | ≤ 2% desirable; adjust batch water if higher |
| Organic impurities | IS 2386 Pt 2 | Colour test — darker than standard is reject |
Reference: IS 383:2016 Tables 5 and 6 (deleterious material limits). The silt limit at 3%/2%/1% is the single most-cited aggregate acceptance number on Indian sites.
Coarse aggregate — nominal sizes, FI/EI, AIV
Coarse aggregate is material mostly retained on 4.75 mm. IS 383 specifies it two ways:
- · Graded — concrete-aggregate nominal sizes (10 mm, 20 mm, 40 mm). Used directly in batching.
- · Single-sized — crusher intermediate. Re-blend before batching; never use as-is for concrete.
Grading envelope — graded coarse
| IS sieve | 10 mm nominal (%) | 20 mm nominal (%) | 40 mm nominal (%) |
|---|---|---|---|
| 63 mm | — | — | 100 |
| 40 mm | 100 | 100 | 95–100 |
| 20 mm | 100 | 95–100 | 30–70 |
| 12.5 mm | — | — | — |
| 10 mm | 85–100 | 25–55 | 10–35 |
| 4.75 mm | 0–20 | 0–10 | 0–5 |
| 2.36 mm | 0–5 | 0–5 | — |
Reference: IS 383:2016 Table 2 (graded coarse aggregate — cumulative % passing by weight).
Mechanical and physical acceptance
| Property | Test method | General concrete | Wearing surface |
|---|---|---|---|
| Aggregate Impact Value (AIV) | IS 2386 Pt 4 §4 | ≤ 30% | ≤ 24% |
| 10% Fines Value (10% FACT) | IS 2386 Pt 4 §4 | ≥ 10 t | ≥ 13 t |
| Los Angeles Abrasion | IS 2386 Pt 4 §5 | ≤ 30% | ≤ 25% |
| Aggregate Crushing Value (ACV) | IS 2386 Pt 4 §3 | ≤ 30% | ≤ 24% |
| Flakiness Index (FI) | IS 2386 Pt 1 §6 | ≤ 25% | ≤ 25% |
| Elongation Index (EI) | IS 2386 Pt 1 §6 | ≤ 25% | ≤ 25% |
| Specific gravity | IS 2386 Pt 3 | 2.6–2.8 typical | 2.6–2.8 typical |
| Water absorption | IS 2386 Pt 3 | ≤ 2% | ≤ 2% |
Reference: IS 383:2016 Tables 7 and 8 (mechanical and physical acceptance). Note the asymmetry between general concrete and wearing surfaces — pavements and floors subject to abrasion have tighter AIV/FACT limits.
FI + EI together — not FI alone. Both flake (thin) and elongated (long) particles break the aggregate skeleton. IS 383 limits each to 25%, but the practical warning is their combined share. A source that is borderline on both produces harsh, unworkable mixes.
AIV vs 10% FACT. AIV is the easier test to run (one hammer pass), so it appears on most QC sheets. 10% FACT correlates with AIV but loads the sample more gently; it is closer to actual in-service behaviour. IS 383 sets a numeric limit on AIV, with 10% FACT as a supplementary check.
Sampling protocol (IS 2386 Pt 1)
Sampling is the step most often skipped or done badly on Indian sites. A grading curve drawn on a contaminated or non-representative sample is worthless — the mix design it feeds will be wrong, and the wrong mix design will produce concrete that meets the wrong target strength.
Collect from the moving stream
IS 2386 Pt 1 · avoid standing-pile bias
Sample at the belt conveyor discharge or the chute outlet — never from a stationary stockpile. Segregation in a stockpile means the bottom-third is oversize and the top-third is fines. One grab from a stockpile is not a sample; it is a guess.
Make a composite of ≥ 5 increments
≥ 5 increments · full cross-section of stream
Take at least 5 increments spaced evenly over the lot. For a 100 m³ lot of fine aggregate that means 5 increments of ~5 kg each, bulked to ~25 kg. Coarse aggregate: 5 increments × ~10 kg = 50 kg bulked.
Reduce with a riffle divider
IS 2386 Pt 1 — quartering is not enough
Pass the bulked sample through a riffle divider (≥ 12 slots, alternating to two pans) until you reach the test portion size. For fine aggregate sieve analysis, the test portion is 1–2 kg. For AIV/10% FACT on coarse, ~3 kg. Quartering by hand loses fines preferentially.
Label with source, lot, date, sampler
IS 2386 Pt 1 — chain of custody
A grading sheet with no source name is not a grading sheet. Date of sampling, supplier lot ID, intended use, sampler initials — all on the label. If a dispute arises ("when did this silt content get this bad?"), the label is the only defence.
Sampling frequency — practical QA cadence
| Aggregate type | Test | Practical minimum frequency |
|---|---|---|
| Fine | Sieve analysis | Every new source or lot change; weekly on continuous supply |
| Fine | Silt content | Daily when stored on site; each shift if rain-exposed |
| Fine | SG, absorption | Each new source / monthly on continuous supply |
| Coarse | Sieve analysis | Each new source; weekly on continuous supply |
| Coarse | FI, EI, AIV, 10% FACT | Each new source; quarterly verification |
| Coarse | SG, absorption | Each new source; monthly on continuous supply |
Reference: contract specifications usually cite a per-100 m³ or per-day frequency; the cadence above is a defensive default when the spec is silent.
Field acceptance — what to actually do at site
You will rarely get a lab AIV result back in time to reject a suspect truck at the gate. The site engineer's job is to filter out the obviously bad deliveries with quick tests, and let the lab confirm or contradict the borderline cases. Three quick tests cover ~80% of the rejection decisions:
Decantation in a 250-ml measuring cylinder. ~50 g dry fine aggregate, water to ~150 ml, shake, settle 3 hr. Silt layer > 3% of sand height = reject or wash. This is a quick field screen — not IS 2386 Pt 2 — but it catches the most common monsoon-related fault (sand fines jump during rain). Back up with the IS sedimentation test before any quality-record decision.
Hand-sieve ~200 g of coarse aggregate on a 20 mm and a 10 mm slotted gauge. Visual estimate of oversize / undersize share. Confirms what the truck driver claims. Does not replace IS 2386 Pt 1 lab sieve, but catches gross misclassifications.
Squeeze a fistful of damp coarse aggregate and release. If fine dust stays on your palm, fines content is high. Look for clay lumps, mica flakes, coal particles. Smell for organic impurities. These are reject indicators that need no lab.
Fine aggregate moisture can be 5–10% during monsoon — that 7% becomes 70 litres of extra water per m³ of concrete. Daily moisture test (IS 2386 Pt 3) on stockpile is the most-overlooked QA step in India.
Worked example — grading report card
Scenario. M30 column pour. Crushed basalt 20 mm nominal, Zone II river sand. Three fine aggregate samples received in the same week from the same supplier. Lab sieve analysis (IS 2386 Pt 1) returns cumulative % passing:
| Sieve | Sample A | Sample B | Sample C | Zone II band | Pass / Fail |
|---|---|---|---|---|---|
| 4.75 mm | 96 | 95 | 92 | 90–100 | A · B · C pass |
| 2.36 mm | 88 | 85 | 78 | 75–100 | A · B · C pass |
| 1.18 mm | 73 | 68 | 55 | 55–90 | A · B · C pass (C at lower band) |
| 600 µm | 48 | 42 | 31 | 35–59 | A · B pass · C marginally low (31 < 35) |
| 300 µm | 20 | 14 | 8 | 8–30 | A · B · C pass (C at lower band) |
| 150 µm | 5 | 3 | 1 | 0–10 | A · B · C pass |
| Silt content | 2.4% | 3.1% | 4.2% | ≤ 2% for M30 | A pass · B · C fail |
Decision. Sample A is acceptable as Zone II. Sample C is borderline on grading and fails on silt (4.2% > 2% for M30) — reject. Sample B fails on silt only. Supplier told: bring silt under 2% or wash before delivery; grading envelope reminder issued.
Why "marginal" still means "reject" for M30. Silt limit tightens from 3% (≤ M25) to 2% (M30 and above) to 1% (marine / aggressive). Silt is a binding-line failure, not a curve-shaped one — you can pass at 1.9% and fail at 2.1% with no middle ground. Borderline is past the line.
Common deviations seen on Indian sites
Silt jumps from 2% to 5–8% in days. Either reject at gate or wash. Batch-water correction alone is insufficient — silt coats aggregate and breaks paste bond.
Mix-design proportions assume 20 mm nominal, not 20+40 mm. If the plant adds 40 mm by accident, the 20 mm bin under-fills and the mix becomes gap-graded and harsh.
Many crusher setups produce M-sand well above the 10% Zone II/III limit. The 18% looks fine but the water demand jumps 8–12% and strength drops 10–15%.
An over-loaded jaw crusher produces FI > 30%. Mixes become harsh, corners honey-comb, cover meter reads inconsistently. Reject the lot or re-crush.
IS 383 was revised in 2016. Older specs may reference deleted limits. Always quote the latest revision; use IS 383:2016 + IS 2386 (latest) on every project.
Field-checklist blocks
A short questionnaire, an inspector's checklist, and three worked examples that show how a delivery of aggregate is read against the IS 383 limits, and resolves into an accept, reject or wash decision before it reaches the batching bin.
Questionnaire — ask when the aggregate is offered at the gate
- Which IS 383 zone or coarse nominal size is this material supplied against, and is the grading being read sieve by sieve against that envelope rather than as a single number?
- What is the silt content by IS 2386 Pt 2, and does it match the limit for the grade being poured: 3% for grades up to M25, 2% above M25, 1% for marine or aggressive exposure?
- Where was the sample taken from, the moving stream or a standing pile, and how many increments went into the composite?
- Was the bulk sample reduced on a riffle divider, and does the label carry the source, the lot, the date and the sampler?
- What is the free moisture of the fine aggregate today, and has the batch water been corrected for it?
- If this is M-sand, what percentage passes 75 µm, and what has that done to the water demand and the admixture dose?
- If this is coarse aggregate, what are the AIV and the 10% FACT for the source, and do the limits tighten to 24% and 13 t because the element is a wearing surface?
- Is this coarse aggregate graded or single-sized, and is it being re-blended before it reaches the batching bin?
Inspector's checklist — what to verify on site
- Zone or nominal size stated on the delivery, and the grading curve checked sieve by sieve against IS 383:2016 Table 2 for coarse aggregate or Table 4 for fine.
- Silt content on record by IS 2386 Pt 2, with the limit applied matched to the grade and the exposure: 3%, 2% or 1%.
- Sampling record: at least 5 increments from the moving stream, riffle-reduced, label carrying source, lot, date and sampler initials.
- Daily moisture test on the fine aggregate stockpile (IS 2386 Pt 3), and the batch water adjusted for the free moisture found.
- Deleterious material screen: clay lumps ≤ 1%, coal and lignite ≤ 1%, mica desirable ≤ 1%, and the colour test for organic impurities.
- Mechanical results on the lab report, not the truck ticket: AIV ≤ 30% (≤ 24% for wearing surfaces), 10% FACT ≥ 10 t (≥ 13 t wearing), FI and EI each ≤ 25%, specific gravity 2.6–2.8, water absorption ≤ 2%.
- Bin discipline: 20 mm and 40 mm not loaded as one truck, and single-sized material re-blended before batching.
- Test frequency against the QA cadence: fine aggregate sieve weekly on continuous supply, silt daily when stored on site, coarse FI, EI, AIV and 10% FACT on each new source with quarterly verification.
What happens if… three worked examples
What happens. For an M30 pour the silt limit is 2%, not the 3% that applies up to M25, and the Zone II envelope puts the 600 µm sieve at 35–59% passing. The three samples read 2.4%, 3.1% and 4.2% silt, and Sample C reads 31% at 600 µm. Take the limits off each result: silt is 0.4, 1.1 and 2.2 percentage points over for A, B and C, and C is also 4 points below the floor at 600 µm, 31 against 35. B and C break even the 3% allowance that applies up to M25, so neither can borrow it. On the rule that a lot can pass at 1.9% and fail at 2.1% with no middle ground, A is over the line too.
Likely outcome. All three lots go back to the supplier to be washed or brought under 2% silt. A is the dangerous one, because its grading curve is clean Zone II and the sheet reads comfortable, so it goes into the bin carrying 2.4% silt, and silt coats the aggregate surface and breaks the paste bond. That is the classic cause of low cube strength with nothing visible in the mix. Subtract the limit from the result before you sign, every time.
Preventive correction. Put the limit next to the result on the acceptance sheet so the subtraction is part of the routine, and run the 250 ml cylinder decantation screen at the gate to catch the bad loads before they are bulked.
What happens. The figure the article gives is 7% free moisture adding 70 litres of extra water per m³ of concrete, which is 10 litres per m³ for every percentage point of moisture. So 8% free moisture puts 80 litres per m³ into the batch and 10% puts 100 litres per m³. On a mix designed at a fixed w/c, the batch water is dosed to the recipe while the sand quietly brings its own.
Likely outcome. The effective w/c climbs and the 28-day strength comes in 5–10% low, while the batch record shows the correct water because the operator did dose to the recipe. Test the moisture on the stockpile every day (IS 2386 Pt 3) and deduct the free water from the batch water. Keep it separate from the silt problem, because washing the sand does not fix the water that went in wet.
Preventive correction. One moisture figure on the plant record for the day, applied to every batch, instead of a correction each operator makes from memory.
What happens. 18% passing 75 µm against the 10% limit for Zone I to III is 8 percentage points over, 1.8 times the limit. The material is not visibly wrong. It is clean manufactured sand out of a crusher, and the sieves you normally check will look plausible while that one fraction sits there.
Likely outcome. The consequence the article records for that material is a water demand up 8–12% and a strength loss of 10–15%, and both of them surface 28 days later with nothing on the file to point at. Check the percentage passing 75 µm on every new crusher lot, before the sand reaches the bin, and either reject the lot or blend it down.
Preventive correction. If it is already in the bin, adjust the water and the admixture to hold the w/c, and expect to be explaining the strength record rather than the mix.
Frequently Asked Questions
What is the silt content limit for fine aggregate under IS 383?
What is the difference between single-sized and graded coarse aggregate?
What is the Aggregate Impact Value (AIV) limit under IS 383?
How many samples of aggregate must be taken per lot?
Can manufactured sand (M-sand) fully replace river sand under IS 383?
Related articles
IS 10262:2019 Mix Design Walkthrough
Proportioning concrete from aggregate properties, target strength and exposure conditions.
IS 456:2000 — Annotated Walkthrough
Clauses the site engineer actually opens: durability, cover, materials, acceptance.
Cube Testing Procedure — IS 516
Sampling, casting, curing and acceptance of concrete cubes against IS 456 Table 11.
Concrete Materials — Field Handbook
Cement, water, admixtures and aggregates — what to check at the gate, what the lab catches.
Amit Haridas
Founder & Proprietor, ConcreteInfo. 25+ years experience in construction QA/QC, concrete technology, RMC plant operations, consulting and technical training across India. For a project-specific aggregate source review, contact amit@concreteinfo.in.
Technical updates, not sales emails
One email when a new article goes out on concrete technology, QA/QC and construction quality. No promotions, and your address is never sold or passed to advertising networks. Subscribing is separate from making an enquiry.