Non-Destructive Testing of Concrete

A spring-loaded plunger is pressed firmly against a smooth, clean concrete surface and released. The plunger rebounds off the surface and the rebound number (R) — typically 10–100 — is read from the scale. The higher the rebound, the harder the surface. A minimum of 10 readings per test location must be taken, discarding any reading that differs from the median by more than 6 units, and averaging the remainder. The test must be conducted on a smooth, dry, non-carbonated surface free from loose aggregate, paint, and surface coatings. Surface orientation affects readings — correction factors must be applied for vertical and downward testing per EN 12504-2. The N-type (standard) Schmidt hammer is used for concrete; the L-type for lightweight or thin elements.

The rebound number is converted to an estimated compressive strength using the manufacturer's calibration chart or a site-specific calibration curve developed from companion core tests. Standard charts assume Ordinary Portland Cement (OPC), 28+ days age, and saturated surface dry (SSD) moisture conditions. Deviations from these conditions introduce significant error: carbonated surfaces give falsely high readings; wet concrete reads 20% low; low cement content concrete reads low. The rebound hammer only tests the surface 30–50 mm zone — it cannot assess the full cross-section or detect internal defects. Never use rebound hammer results alone for a structural strength assessment — always calibrate with cores per AS 3600 Appendix B6.

An ultrasonic pulse (50–500 kHz) is generated by a transmitter transducer pressed against one concrete surface. The pulse travels through the concrete and is received by a receiver transducer on the same or opposite surface. The instrument measures the transit time (μs) and the operator measures the path length (mm) — pulse velocity (m/s or km/s) is calculated as path length ÷ transit time. Three probe arrangements are used: direct (opposite faces — most accurate), semi-direct (adjacent faces), and indirect/surface (same face — least accurate, only when one face is inaccessible). Coupling gel must be applied to both transducers. Minimum 5 readings per test location. Path length must be accurately measured — errors in path length directly error velocity and quality classification.

UPV quality grading is universally referenced from the table originally published by Leslie and Cheesman (1949) and subsequently adopted in IS 13311, BS 1881, and VicRoads TN-061. The grading assumes direct transmission on sound, well-cured concrete at ambient temperature. Dense, well-compacted, crack-free concrete typically produces velocities above 4.5 km/s. Velocities below 3.0 km/s indicate poor quality, significant voids or cracking, or very low cement content. Reinforcing steel bars aligned with the pulse path increase apparent velocity — test paths must be oriented to avoid bar alignment wherever possible. Moisture increases velocity by up to 5%.

SONREB (SONic REBound) combines the UPV test and the rebound hammer test at the same test location. The two methods are complementary because their errors are partially offsetting — moisture increases UPV but decreases rebound number; carbonation increases rebound but does not affect UPV; cement content affects both but in different proportions. By using both measurements simultaneously in a combined regression equation, the uncertainty in strength estimation is reduced from ± 20–30% (single method) to approximately ± 10–15%. RILEM Technical Committee 43-CND published the most widely referenced SONREB correlations and calculation nomographs. AS 3600 Appendix B6 notes that "combined non-destructive techniques have been found to substantially improve the order of accuracy of the estimated values."

Conduct the rebound hammer test (minimum 10 readings, average) and UPV test (direct transmission if possible) at the same marked test location on the structure. Apply the SONREB equation: f'c = a × V^b × R^c where V is pulse velocity (km/s), R is mean rebound number, and a, b, c are empirically derived calibration constants. Standard published constants (RILEM) give a useful first estimate, but site-specific calibration using a minimum of 3 concrete cores from the structure being assessed — testing cores for UPV, rebound, and compressive strength — dramatically improves accuracy. Never rely on published constants alone for a formal structural assessment or regulatory compliance judgement.

A copper/copper sulphate (CSE) or silver/silver chloride half-cell electrode is pressed against the concrete surface (through a wet sponge to ensure electrical contact) while a high-impedance voltmeter measures the electrical potential difference between the half-cell and the reinforcing steel (connected via an electrical lead to an exposed bar or drilled access point). Potential readings are taken on a grid pattern — typically 200–300 mm centres — and plotted as a potential contour map. The map reveals zones of active corrosion (strongly negative potential), passive steel (less negative), and zones of transition. Testing requires that the concrete surface be wetted to establish adequate electrical conductivity. Dry or carbonated concrete requires pre-wetting for a minimum of 4 hours before testing.

Potential readings are interpreted against the ASTM C876 threshold values (CSE reference electrode). A potential more negative than –350 mV CSE indicates greater than 90% probability of active reinforcement corrosion at that location. Between –200 mV and –350 mV, corrosion activity is uncertain. Less negative than –200 mV indicates a greater than 90% probability that no active corrosion is occurring. Importantly, the test identifies probability of corrosion, not its severity — it cannot determine how much section loss has occurred. Follow-up investigation using covermeter, carbonation testing, and chloride content sampling is needed to determine the cause and severity of any corrosion identified. Not valid for epoxy-coated or galvanised reinforcement.

The covermeter uses electromagnetic induction — a coil in the probe generates a magnetic field that is disturbed by ferromagnetic steel reinforcement. The disturbance is measured and converted to a cover depth reading. Modern digital covermeters also provide an estimate of bar diameter. The probe is passed over the concrete surface in a grid or line-scan pattern. A signal peak indicates a bar crossing — the cover depth displayed is the distance from the probe face to the nearest steel surface. Survey grids of 100–200 mm are used for mapping. An initial orientation scan (parallel and perpendicular sweeps) establishes the bar layout before close-interval measurements are taken. Results must be calibrated against the actual bar diameter — an incorrect bar diameter input introduces error in the cover reading.

Cover readings are compared against the minimum cover requirements of AS 3600:2018 Table 4.10.3.2 for the specified exposure classification. For interior members (A1): minimum cover 20 mm. For near-coastal (B1): 35 mm. For coastal (B2): 40 mm. For marine splash (C1): 50 mm. Covermeter surveys on existing structures often reveal cover deficiencies in areas where formwork shifted during pour or where vibration displaced reinforcement. Results showing cover below the design minimum trigger a durability assessment — low cover areas are at significantly elevated risk of carbonation-induced or chloride-induced corrosion. Accuracy is reduced where bars are closely spaced (< 75 mm), large aggregate is present, or metallic embedded items are nearby.

Atmospheric CO₂ reacts with calcium hydroxide in the concrete to form calcium carbonate — a process called carbonation — which reduces the pH of the concrete pore solution from ~13 to below 9. When pH falls below approximately 9.5, the passive oxide film protecting the reinforcing steel breaks down, enabling corrosion. Carbonation depth is measured by spraying a 1% phenolphthalein indicator solution onto a freshly broken or drilled concrete face. Carbonated concrete (pH < 9.5) remains colourless; uncarbonated concrete (pH > 9.5) turns bright pink/purple. The depth of the colourless zone is measured with a calliper to the nearest millimetre. A minimum of 3 measurements per location, averaged. The test requires a fresh break — do not test a surface exposed to air for more than 2 minutes before spraying.

The carbonation depth is compared with the actual concrete cover to reinforcement (from covermeter survey). If the carbonation depth equals or exceeds the cover depth, the reinforcement is at risk of corrosion initiation. Carbonation progresses approximately in proportion to the square root of time — a structure that has carbonated 10 mm in 10 years will reach approximately 20 mm in 40 years. Typical carbonation rates: well-cured, dense N32+ concrete: 0.5–1 mm/year; poorly cured or low-grade concrete: 2–4 mm/year. Carbonation rate is accelerated by low relative humidity (50–65% RH is the most aggressive range), high porosity, low cement content, and incomplete curing in 2026.

GPR transmits short electromagnetic radar pulses (typically 1–2.6 GHz for concrete scanning) into the structure and records the time and amplitude of reflections from interfaces between materials of different dielectric properties — steel bars, voids, delaminations, water, and layer boundaries all produce distinct reflections. The data is displayed as a 2D B-scan (cross-section profile) or processed into a 3D volume. GPR is the only NDT method capable of producing a complete 3D map of all reinforcement (including bar depth, spacing, and layout), locating post-tensioning ducts, identifying voids and honeycombing within the full slab or wall depth, and detecting delamination at overlay interfaces — all in real time without drilling. Essential before any saw-cutting or core drilling in prestressed or post-tensioned concrete.

GPR data interpretation requires significant expertise — misidentification of reflections is common among untrained operators. Key limitations: penetration depth in concrete is limited to approximately 400–600 mm in dry concrete, much less in saturated or high-chloride concrete (dielectric absorption increases dramatically); closely spaced reinforcement (< 100 mm) creates overlapping hyperbola patterns that obscure deeper features; GPR cannot reliably detect chloride content, concrete strength, or corrosion extent. A GSSI, IDS Georadar, or equivalent calibrated concrete scanning system operated by a Level 2 NDT technician is required for a reliable result. Raw GPR data output must always be accompanied by an interpreted report from the operator — never accept a raw scan without interpretation.

A metal insert (disc or bolt) is either cast into the fresh concrete (LOK-Test) before placement or drilled and installed into hardened concrete (CAPO-Test). A pulling rig is attached to the insert and a measured tensile force is applied until the insert pulls out a frustum-shaped concrete cone. The measured pull-out force is directly correlated to compressive strength through a calibration relationship. The LOK-Test insert is embedded at 25 mm depth; the CAPO-Test insert (drilled to 25 mm with a 55 mm counter-bore) closely replicates LOK geometry in existing structures. A minimum of 4–6 pull-out tests per assessment zone are required. The test is considered semi-destructive as it creates a small surface void, but this is easily patched with repair mortar.

The pull-out test has a significantly better strength correlation than rebound hammer or UPV alone — the standard deviation of the pull-out force to compressive strength relationship is typically ± 8–12% compared to ± 20–30% for single surface methods. It is particularly valuable for: early age strength verification (formwork stripping decisions — test at 12, 24, 48 hours after pour); cold weather concreting where cylinder tests may not reflect in-situ conditions; verification of concrete placed in inaccessible locations; and confirming that a failed cylinder result was a test artefact rather than a structural deficiency. The CAPO-Test is widely used in Australia by specialist concrete investigation firms for post-construction strength assessment of existing structures.