Reference card — verified IS codes
Verified 2026-09-14 against ~/Standards library + BIS portal. ConcreteInfo reference card.
Codes & standards referenced
Schmidt hammer calibration, test grid (≥16 readings), angle and atmospheric corrections. Limits applicability to surface layer (~30 mm depth) and relational comparisons only.
Method for measuring UPV through concrete. Velocity grading: > 4.5 km/s good, 3.5–4.5 doubtful, < 3.5 poor. Confirm the latest published section number via the BIS catalogue.
Procedure for extracting, preparing, and testing cores for compressive strength. Critical when cubes and structure disagree. Pair with ASTM C42 / BS 1881:207 cross-references for international project reports.
Useful cross-reference for combined NDT correlation curves. Cite IS standards in Indian project reports.
1 — Rebound hammer (IS 516 Pt 5/Sec 4)
The Schmidt hammer measures surface hardness by firing a spring-loaded mass against concrete and reading the rebound distance on a 0–100 scale. It tests only the outer ~30 mm of concrete — carbonated surface gives misleadingly high readings.
The values below are taken from the Original Schmidt Type N manufacturer's published correlation curve, assuming horizontal impact on standard 150 mm cubes at 28 days, on an uncarbonated, well-prepared surface. Real-world readings vary ±20–30% from this curve. Corrections must be applied for:
- Hammer orientation — vertical-down reads ~5–7 points higher than horizontal; vertical-up (soffit) reads ~5–7 points lower. Use the angle-correction chart from the hammer manual.
- Concrete age & curing — 7-day readings run ~15% lower than 28-day; early-age tests overstate later strength gain.
- Carbonation depth — on a 5+ year-old structure, uncorrected rebound numbers overstate strength by 30–50% (see warning below).
- Surface preparation — grind off laitance; rough formwork marks and honeycombing give false-low readings.
- Aggregate hardness — basalt and quartzite read higher than limestone for the same paste strength.
IS 516 Pt 5/Sec 4 forbids using rebound hammer alone for acceptance. Build a project-specific curve using trial mixes + at least 3 paired (cube + rebound) data points before relying on this table for any decision beyond uniformity assessment.
| Average rebound number (R) | Indicative compressive strength (MPa, horizontal impact, 28-day) | Quality interpretation |
|---|---|---|
| > 50 | ≥ 50 (typical 45–55) | Very high strength — typical M50 and above; verify with cores before any acceptance decision |
| 40 – 50 | 32 – 50 (typical M40 grade) | High-strength structural concrete (typical M40) |
| 30 – 40 | 22 – 40 (typical M25–M35) | Normal structural concrete (typical M25–M30) |
| 20 – 30 | 14 – 28 (typical M15–M20) | Low strength or suspect quality — verify with cores before any decision |
| < 20 | < 18 (typically < M15) | Poor — immediate investigation required (cores + UPV + structural review) |
Source: Original Schmidt Type N manufacturer's correlation curve, horizontal impact on 150 mm cubes at 28 days, uncarbonated surface. References: Proceq Operating Instructions — Original Schmidt Concrete Test Hammer; Bungey, The Testing of Concrete in Structures, 4th ed.; IS 516 Pt 5/Sec 4 calibration clause. Apply angle, age, and carbonation corrections before use. Build a project-specific correlation per IS 516 Pt 5/Sec 4 for any acceptance decision.
Concrete surface carbonates over time, increasing hardness. On a 5+ year-old structure, uncorrected rebound readings overstate strength by 30–50%. Either grind 5 mm off the test point, or apply a correction factor based on phenolphthalein-spray depth-of-carbonation measurement.
Field protocol (per IS 516 Pt 5/Sec 4)
- · Test grid: minimum 16 readings in a 300 × 300 mm square, discard outliers (±7 from median).
- · Surface must be smooth — grind if necessary. Avoid honeycomb, laitance, and rough formwork marks.
- · Hold hammer perpendicular to surface (90° ± 5°) — angle corrections are mandatory.
- · Calibrate daily on the test anvil supplied by manufacturer.
- · Atmospheric correction: -3 to +5% per 10°C deviation from 27°C reference.
2 — Ultrasonic pulse velocity (IS 516 series)
UPV sends a 20–150 kHz pulse from a transmitter and times the arrival at the receiver. Velocity in well-compacted, dense concrete is 4.0–5.5 km/s; lower velocity indicates internal cracks, voids, or honeycombing.
UPV grading (IS 516 series)
Dense, well-compacted, homogeneous. No defects expected.
Some voids / minor cracking. Further investigation warranted.
Significant defects. Core sampling and structural review required.
Velocity in km/s. Values indicative — use combined-method assessment per IS 516.
Probe configuration
Both probes on opposite faces. Most accurate. Required for beams, columns. Path length must be measured accurately.
Probes on adjacent faces (90° apart). Used for slabs on inaccessible soffit. Slightly less accurate.
Both probes on same face. Detects only outer ~50 mm. Used for one-sided access only (retaining walls). Less reliable.
3 — Core sampling (IS 516 Pt 5)
The core test is the only direct measurement of in-place strength in the disputed region. IS 516 Pt 5 / IS 1199 Annex specify 100 mm nominal diameter cores (75 mm acceptable where reinforcement congestion prevents 100 mm), tested as-moulded (L/D = 2.0 ± 0.05) or capped.
Core test decision flow
- Cubes fail (< fck − 3 on any individual result, or the mean of four < fck + 0.825σ) at 28 days.
- Verify cubes: re-test using capped cubes, retained sample, or witness cubes.
- If cubes still fail — independently sample fresh concrete for trial mixes.
- Run combined NDT (rebound + UPV) on suspected structural region.
- If NDT suggests in-place strength is low, schedule core extraction.
- Extract ≥ 3 cores per zone. Core size 100 mm (avoid bars; GPR-scan first).
- Test cores at 28-day equivalent (if tested earlier, apply 0.85 correction).
- Apply strength reduction factor for L/D ratio (0.87 for L/D = 1.0).
- Compare core strength to fck + 3 MPa for acceptance (Indian Engineering Practice).
Every core leaves a hole. The structure must be restored with epoxy-bonded patch using parent-equivalent or higher-grade concrete. Failure to restore properly turns the inspection into a defect.
Other methods — when they fit
Measures force to extract an embedded insert. Strong correlation with compressive strength (~±10%). Excellent for in-place assessment. Less common in India.
Steel pin driven into concrete; depth of penetration correlates to strength. Quick but destructive and inconsistent. Rare in Indian practice.
Estimates rebar corrosion probability (not strength). Used in durability audits of bridges, coastal structures.
Locates rebars, voids, delamination before coring. Reduces "oops, hit a bar" damage to structure.
Common pitfalls on Indian sites
One impact at one point is meaningless. Read 16+ on a 300 mm grid, discard outliers, report the median. IS 516 Pt 5/Sec 4 forbids single-impact acceptance.
Carbonation overstates strength; saturated concrete understates UPV velocity. Always note the surface condition and time of test in the report.
The correlation curve supplied with a Schmidt hammer is generic. For acceptance work, build a project-specific curve using trial mixes + tested cores (3+ pairs).
Hitting a rebar invalidates the core. Always scan with cover meter + GPR before coring. Reposition 80 mm if bar detected.
Field-checklist blocks
A short questionnaire, an inspector's checklist, and three worked examples that show how a rebound number, a pulse velocity and a core reading combine into a defensible verdict on the concrete in the structure.
Questionnaire — ask before the first reading is taken and before any hole is drilled
- How old is the concrete in the element we are testing, and is the surface carbonated? On a 5+ year-old structure, uncorrected rebound readings overstate strength by 30–50%.
- Are we testing for uniformity or for acceptance? IS 516 Pt 5/Sec 4 forbids using rebound hammer results alone to accept or reject concrete, so an acceptance decision needs two methods minimum and cores where the dispute is.
- Do we have a project-specific rebound-to-strength curve built from trial mixes and at least 3 paired cube-and-rebound data points, or are we leaning on the manufacturer's generic curve?
- What is the UPV path we can achieve: direct, semi-direct, or indirect? Indirect detects only the outer ~50 mm, so it cannot be used to assess a beam or column.
- Have we scanned with a cover meter and GPR before choosing core locations, so that we are not drilling straight into a bar?
- What is the core's length-to-diameter ratio going to be, and who restores the hole once it is out?
Inspector's checklist — what to verify on site
- Rebound grid: at least 16 readings in a 300 × 300 mm square, outliers at ±7 from the median discarded, and the median reported rather than the average of everything.
- Hammer condition and geometry: calibrated on the manufacturer's anvil that day, held perpendicular at 90° ± 5°, and the angle correction applied (vertical-down reads ~5–7 points higher than horizontal, vertical-up ~5–7 points lower).
- Surface and condition record: laitance ground off, no honeycomb or rough formwork marks, concrete age noted, carbonation depth measured by phenolphthalein spray where the structure is 5+ years old.
- Atmospheric correction: −3 to +5% per 10 °C deviation from the 27 °C reference applied to the readings.
- UPV record: probe configuration (direct, semi-direct or indirect), measured path length, the velocity in km/s, and which band it falls in (above 4.5 excellent, 3.5–4.5 doubtful, below 3.5 poor).
- Cover meter and GPR scan before coring, with core positions moved 80 mm if a bar is detected.
- Core log: diameter 100 mm nominal (75 mm only where congestion prevents 100 mm), L/D ratio, the 0.87 reduction if L/D = 1.0, and the 0.85 correction if the core was tested before the 28-day equivalent.
- Restoration: every hole closed with an epoxy-bonded patch of parent-equivalent or higher-grade concrete, and the restoration recorded against the core location.
What happens if… three worked examples
What happens. A median of 42 sits in the 40–50 band, which the manufacturer's curve reads as 32–50 MPa and calls typical M40. On a structure older than 5 years, uncorrected rebound overstates strength by 30–50%, so the true value could be as low as 42 ÷ 1.5 = 28.0 MPa or as high as 42 ÷ 1.3 = 32.3 MPa. The number that looked like M40 is anywhere between M25 and M30.
Likely outcome. No acceptance on that basis. You either grind 5 mm off the test point or apply a correction based on the phenolphthalein depth-of-carbonation measurement, then re-read. Until that is done, the column is a candidate for cores and UPV, not a sign-off.
Preventive correction. On any structure 5+ years old, measure carbonation depth before the first hammer blow, and write the correction into the report next to the raw readings.
What happens. The two methods disagree. Rebound says the surface is hard and strong: 45 falls in the 40–50 band, indicative 32–50 MPa, typical M40. UPV says the interior is doubtful, because 3.8 km/s sits in the 3.5–4.5 km/s doubtful band. That combination is the signature of a sound-looking surface over voids or honeycombing, or a carbonated surface over a compromised interior.
Likely outcome. The conflict goes to cores, at least 3 per zone, and the structural review follows. You do not average the two and you do not pick the favourable one. IS 516 Pt 5/Sec 4 is explicit that rebound results alone shall not be used to accept or reject concrete, and a high rebound with a low UPV is a defect signature until cores say otherwise.
Preventive correction. Use UPV as the map and rebound as the ranking. Where they disagree, the map wins and you go to the core.
What happens. Drilling into the bar invalidates the core and does permanent damage to the steel. The position moves 80 mm, the core comes out at the specified 100 mm diameter with L/D = 2.0 ± 0.05, and the test is valid. Had the core been taken short at L/D = 1.0, the 0.87 strength reduction applies, which takes 13% off the reading before any comparison is made.
Likely outcome. The extra 80 mm is cheap. A voided core means re-drilling, another hole to restore, and a schedule that has moved for nothing. Whichever way it goes, the hole gets closed with an epoxy-bonded patch of parent-equivalent or higher-grade concrete, because a core that is not restored properly turns an inspection into a defect.
Preventive correction. Scan before you drill, every time. Set out the core locations on the cover meter and GPR record first, and mark them on the drawing so the next engineer does not drill the same spot.
Frequently Asked Questions
Related articles
Amit Haridas
Founder & Proprietor, ConcreteInfo. Has performed combined NDT assessments (rebound + UPV + core) on RMC plant acceptance failures, bridge durability audits, and post-fire damage assessments across India. IS-codes-aligned reporting; results survive cross-examination in arbitration.
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.