Codes referenced
"The amount of added water shall be adjusted to compensate for any observed variations in the moisture contents in the aggregates. Suitable adjustments should also be made in masses of the aggregates due to this variation (see IS 456)."
"When the assessment of moisture content is carried out continuously using a suitable instrument a system shall be provided to indicate the necessary changes in added water for changes in moisture content."
Oven-drying procedure for aggregate moisture content (well-ventilated oven, 100–110 °C to constant mass) and water-absorption by the SSD cloth-drying method.
The oven-dry reference that RMC plants use alongside IS 2386 Pt 3 for daily probe verification. Identical oven temperature, identical mass-loss calculation.
Production-side reference for moisture accounting and for the rule that the producer shall adjust batch water for aggregate moisture.
Aggregate acceptance. IS 383's grading envelope and aggregate classification are what make the moisture correction meaningful — wrong grading turns a 6% sand moisture into a 10% effective water swing.
Why moisture correction matters
A batcher that loads 750 kg of "dry" sand actually loads 750 kg of wet sand, of which 6% (45 kg) is free water sitting on the particle surface. If the controller treats that 750 kg as dry mass and adds the full 190 kg of mix water on top, the batch receives 190 + 45 = 235 kg of effective water against the 320 kg of cement — an effective w/c of 0.73, not the design 0.45. Slump jumps 30–40 mm above target; 28-day strength drops by 8–12 MPa on a 40 MPa mix.
The reverse error is just as bad. If the controller over-corrects (assumes 10% sand moisture when actual is 4%), it cuts 6% too much water — concrete arrives at site stiff, workability is poor, the temptation to add water on site to "fix" it is irresistible. Both directions erode the strength that the mix design paid for.
Three things control the magnitude of the swing: (1) the moisture content itself, (2) the batcher's accounting (whether it reads the live SM% into the corrected weight and water), and (3) the freshness of the moisture reading relative to actual bin condition. The first is material; the second and third are operational discipline.
The SM / SSD / bone-dry trichotomy
Three aggregate moisture states — each with a different water correction. IS 2386 Pt 3 defines the laboratory methods; the batcher applies the rule that follows from the state.
| Aggregate state | What it means | Water correction |
|---|---|---|
| Bone-dry (OD) | All pores empty — particle has zero internal water and zero surface water | Add absorption% × dry mass of water (to satisfy internal porosity) |
| Saturated surface-dry (SSD) | All pores full, but surface film removed — particle contributes zero free water to the mix | No correction — aggregate is at "neutral" state |
| Wet (above SSD) | All pores full, plus free surface water — particle donates water to the mix | Subtract SM% × wet mass from added water; weigh out wet mass |
Reference: IS 2386 (Part 3):1963 — water absorption and surface moisture determination. Real bin aggregate at a working RMC plant is almost always in the third row.
Worked example — M30 mix, post-monsoon shift
A 1 m³ M30 mix design calls for: cement 380 kg, fly-ash 60 kg, Zone-II sand 720 kg (dry), 20 mm crushed 660 kg (dry), 10 mm crushed 440 kg (dry), water 198 kg, PCE admixture 3.3 kg. Target w/c = 0.42. The bin probe at 8:30 a.m. after a wet night reads: sand SM = 6.0%, 20 mm SM = 1.5%, 10 mm SM = 2.5%.
| Component | Dry recipe (kg/m³) | SM (%) | Wet mass to load (kg) | Free water donated (kg) |
|---|---|---|---|---|
| Sand (Zone II) | 720 | 6.0 | 720 × 1.06 = 763.2 | 43.2 |
| 20 mm crushed | 660 | 1.5 | 660 × 1.015 = 669.9 | 9.9 |
| 10 mm crushed | 440 | 2.5 | 440 × 1.025 = 451.0 | 11.0 |
| Subtotal | 1820 | — | 1884.1 | 64.1 |
Net added water to hit design 198 kg = 198 − 64.1 = 133.9 kg. If the operator skips the correction, the batch receives 198 + 64.1 = 262.1 kg against 440 kg of cementitious, i.e. w/c = 0.60 — almost 43% over target. Expected 28-day strength on the uncorrected batch: ~25–28 MPa, well below the 30 MPa fck.
Reference: IS 4926:2003 §10.2.2.2 (water adjustment for aggregate moisture). The opposite under-correction (over-deducting) gives concrete too stiff to pump.
How a moisture probe works — and why it drifts
The two dominant probe technologies in Indian RMC plants:
- Capacitance / RF admittance probe. A pair of electrodes in the bin wall measures the dielectric constant of the aggregate between them. Water has dielectric ≈ 80; dry rock ≈ 4–6. The reading is a linearised function of the volumetric moisture. Cheaper, drifts with material density change (e.g. crushed vs river sand), fouled by cement or clay dust clinging to the electrode face.
- Microwave resonator probe. Measures the shift in resonance frequency with water content. Less sensitive to material density, more accurate across aggregate type changes, ~3× cost. The choice on premium plants.
The probe gives a continuous number — but it is a proxy for the moisture of the aggregate that has just left the bin, not the aggregate that is in the bin right now. Three drift mechanisms are routine:
Cement/fines dust films the electrode face. Probe reads the film, not the aggregate. Symptom: probe reading flat at one value while oven value drifts.
River sand vs crushed sand, basalt vs granite — same probe calibration, different dielectric baseline. Probe reading jumps without actual moisture change.
When bin level drops below the probe face, the probe reads mostly air. Controller either holds the last good value or throws an alarm. Either way, the batch from that fraction is uncorrected.
Daily verification — the oven-dry check
The probe is fast; the oven is right. The plant discipline is to verify probe against oven, twice per shift, per fraction. IS 2386 Pt 3 specifies the oven procedure.
Pull a representative grab
IS 2386 Pt 3 · grab at bin discharge
- · Take ~500 g from the discharge stream of the bin, mid-batch. Do not sample from the top of the stockpile or the surface of the bin — both give misleading readings.
- · Weigh immediately on a 0.1 g balance to get the wet mass Mw. Cover the sample pan to prevent further evaporation before oven loading.
Oven-dry to constant mass
IS 2386 Pt 3 · ASTM C566 · 100-110 °C
- · Place the sample in a ventilated oven at 100–110 °C. Maintain for at least 24 h for 20 mm aggregate, less for sand, until two successive weighings at 1 h interval differ by less than 0.1% of the sample mass.
- · Cool in a desiccator to room temperature, then weigh to get the oven-dry mass Md.
Compute moisture & compare
SM% = (Mw − Md) / Md × 100
- · Calculate SM% and compare against the probe reading logged at the same instant.
- · Deviation ≤ 1.0 percentage point: probe is in spec; log both values, update the controller with the oven value.
- · Deviation > 1.0 percentage point on one check: re-sample, re-check. If it persists: clean probe, re-calibrate, or replace. Do not load any batch with an unverified probe reading until the check is closed.
Batcher integration — manual, semi-auto, fully auto
Operator types SM% into the console from a separate lab reading. The console computes corrected weights and water. Risk: console value drifts from lab value, no automatic verification. Acceptable for low-volume plants but not for steady 60 m³/h production.
Probe reading auto-feeds the console; operator verifies against the morning oven check and overrides if drift > 1%. Best risk-control/value on a mid-sized plant. Most plants in India run this mode.
Probe → controller → batcher → ticket, all digital. Daily oven check by QA writes back to the controller as a calibration offset. Requires reliable probes and disciplined QA — when it works, it removes the largest single source of human error.
Common deviations seen on Indian sites
Probe installed at commissioning, never re-zeroed against an oven. Three months later the plant is systematically over-batching water — every batch fails 7-day strength, and the root cause is invisible because the console is "working".
Mix design is keyed to SSD aggregate; controller does not correct for moisture because "we use SSD aggregate". Real aggregate is wet, not SSD. Same 1.5% silent over-watering that drops strength.
Probe verified only at 9 a.m. By 4 p.m. after a sunny spell the sand SM has dropped from 7% to 3%. Controller still applies the morning correction. Net: under-watering in afternoon batches, site complaints about stiff concrete.
Probe reads 5.0%, oven reads 7.0%. Operator trusts the probe because "the console is automated". A probe reading that disagrees with the oven, persistently, is a probe that needs cleaning — not a probe that overrides the oven.
Field-checklist blocks
A short questionnaire, an inspector's checklist, and three worked examples that show how the batcher's moisture correction resolves into a decision: load the bin, hold the batch, or go and clean the probe.
Questionnaire — asked at the batch panel, before the corrected batch is released
- What SM% is in the controller for each fraction right now, and when was it last checked against the oven? The number in the console is only as good as the last verification behind it.
- Has the probe been verified against the oven twice this shift, for each fraction? Twice per shift per fraction is the discipline. Once at shift start is not the same thing.
- Does the controller do both legs of the correction? Aggregate mass up, added water down by the same kilogram, per IS 4926:2003 §10.2.2.2. One leg on its own is not a correction.
- Which mode is the batcher in, and who signs the value that goes in? Manual, semi-auto or fully auto. Each one has a different failure mode, and each one needs a named person behind the number.
- Since the last check, has it rained, has fresh aggregate been delivered, or has the bin been topped up? All three move the moisture, and all three mean another check.
- Is the bin level above the probe face? Below the probe the reading is mostly air, and the controller either holds a stale value or alarms. Either way that fraction is uncorrected.
- Has anybody added water in the truck drum since batching? Not without the purchaser's request and signature, per IS 4926:2003 §10.2.1.6. It is the same w/c question, asked twice.
Inspector's checklist — what to verify on site
- Representative grab: about 500 g from the bin discharge stream, mid-batch, weighed immediately on a 0.1 g balance for the wet mass. Not from the top of the stockpile and not off the surface of the bin, both of which lie.
- Oven to constant mass, per IS 2386 (Part 3):1963: ventilated oven at 100–110 °C, at least 24 h for 20 mm aggregate and less for sand, until two successive weighings at a 1 h interval differ by less than 0.1% of the sample mass. Cool in a desiccator before the dry mass.
- Moisture computed and compared: SM% = (Mw − Md) / Md × 100, against the probe reading logged at the same instant.
- Deviation closed out before the next batch: ≤ 1.0 percentage point means log both values and update the controller with the oven value. Above 1.0 point, re-sample, re-check, then clean the probe face, re-calibrate or replace. No batch is loaded on an unverified reading until the check is closed.
- The batch record carries the corrected figures, not just the dry recipe: the wet aggregate masses weighed out and the reduced added water.
- Check log per fraction per shift, with the extra checks the article calls for: after rain, after a fresh delivery, and whenever the bin level crosses the probe zone.
- Mode discipline: in manual mode the operator's SM% is matched to the oven reading before the batch; in semi-auto any override is recorded when drift exceeds 1%; in fully auto the oven value is written back to the controller as a calibration offset by QA.
- No water in the drum after batching without the purchaser's request and signature, per IS 4926:2003 §10.2.1.6.
What happens if… three worked examples
What happens. The M30 recipe per m³ is 720 kg dry sand, 660 kg dry 20 mm, 440 kg dry 10 mm, 198 kg water and 380 kg cement plus 60 kg fly ash. The probe reads sand 6.0%, 20 mm 1.5%, 10 mm 2.5%. Free water comes off the aggregate at 720 × 0.06 = 43.2 kg, 660 × 0.015 = 9.9 kg and 440 × 0.025 = 11.0 kg, so 64.1 kg in total. Wet masses to load are 763.2, 669.9 and 451.0 kg, and the water that actually goes in is 198 − 64.1 = 133.9 kg.
Likely outcome. The batch keeps the design w/c of 0.42 and the mix behaves like the design. Skip the correction and the same batch takes 198 + 64.1 = 262.1 kg against 440 kg of cementitious material, a w/c of 0.60 against a 0.42 target, roughly 43% over. That batch comes back at about 25–28 MPa at 28 days, under the 30 MPa fck.
What happens. The operator trusts the console because it is automated. On the 720 kg of dry sand above, the controller deducts 720 × 0.05 = 36 kg of free water when the truth is 720 × 0.07 = 50.4 kg, so 14.4 kg of water goes into the mix that nobody asked for. The probe is not slightly wrong here, it is 2.0 percentage points out, and that is a reading that must not be trusted.
Likely outcome. Every batch runs wet, slump drifts high, and the strength the mix design paid for quietly goes. The probe face needs cleaning and re-calibration, and until the check is closed no batch is loaded on that reading. A probe that disagrees with the oven needs cleaning, not trust.
What happens. The 9 a.m. check is done and the plant moves on. By 4 p.m. the sand SM has dropped from 7% to 3%. On 720 kg of dry sand the controller is still deducting 720 × 0.07 = 50.4 kg when the aggregate is only carrying 720 × 0.03 = 21.6 kg, so 28.8 kg of water is missing from the batch and the morning correction is still sitting in the console.
Likely outcome. Afternoon batches arrive stiff, the site calls it a quality complaint, and the temptation on site is to add water, which takes you straight back to the w/c problem. The fix is not more water, it is a second check mid-shift and after any weather change.
Frequently Asked Questions
How often should aggregate moisture be tested at an RMC plant?
What is the acceptable moisture probe deviation from the oven value?
Can moisture correction be skipped if the aggregate is at saturated surface dry (SSD) condition?
Related articles
RMC Plant Quality Management — Five Control Zones
Where moisture correction sits inside the broader plant QA chain.
Water-Cement Ratio Explained — Strength, Durability & IS 456
Why every kilogram of uncorrected water silently moves strength and durability.
Mix Design Masterclass — IS 10262 Walkthrough
Where the dry-weight recipe — the "input side" of the moisture correction — is actually generated.
Amit Haridas
Founder & Proprietor, ConcreteInfo. 25+ years experience in construction QA/QC, concrete technology, and RMC plant operations. NRMCA Certified Trainer (USA) and ISO Lead Auditor. For moisture-probe calibration audits, mix-design moisture economics, or batch controller setup 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.