The method is used to assess internal uniformity, detect voids, estimate crack depth and monitor concrete deterioration without damaging the structure. This article explains the measurement principle, transducer arrangements, field procedure, how to read velocity values and the limitations to know before ordering a test.

What Does a UPV Test Measure and How Does It Work?

UPV measures the velocity of a compressional wave (P-wave) traveling through concrete. A pulse is sent from a transmitter through the concrete to a receiver; the instrument records the travel time (transit time) and the velocity is calculated from the formula

Pulse velocity (V)  =  transmitter–receiver path length (L)  ÷  transit time (T)

The result is usually reported in kilometers per second (km/s). For example, a path length of 250 mm with a reading of 62.5 microseconds (µs) gives a velocity of 4.0 km/s. Dense, defect-free concrete gives a high velocity because the pulse travels quickly through solids, while voids or air-filled cracks force the pulse to detour, so the travel time increases and the calculated velocity drops.

UPV principle: ultrasonic pulse through dense concrete compared with concrete containing a void, and the pulse arrival times Two panels show cross-sections of a concrete column with transmitter T and receiver R placed on opposite faces in the direct arrangement; the distance between transducers is L. In the left panel the concrete is dense and the pulse travels in a straight line. In the right panel there is a void or honeycombing in the middle of the section; the pulse cannot cross air, so it must travel around the edge of the void. Below each panel is the received signal on the same time axis: in the left panel the pulse arrives early at time t1, in the right panel it arrives later at t2, so the velocity is lower. The bottom strip shows the formula V equals L divided by t per ASTM C597, with the example 250 millimeters divided by 62.5 microseconds equals 4.0 kilometers per second, and a note that couplant must be applied to the transducers. UPV: pulse detours void = longer time = lower velocity Not to scale Dense concrete T R Transmitter Receiver L (same in both) Time → t₁ early High V Void / honeycomb Void T R Transmitter Receiver L (same in both) Time → t₂ late Low V 1 V = L / t ASTM C597 2 250 mm ÷ 62.5 µs = 4.0 km/s 3 Apply couplant to probes Measure L accurately
Left: the pulse travels straight through dense concrete, high velocity — Right: the pulse must detour around a void/crack, longer travel time, lower velocity

Which Transducer Arrangements Can Be Used?

There are 3 transducer arrangements, chosen according to how many faces of the member are accessible. All three are defined in BS EN 12504-4, while ASTM C597 focuses on direct transmission, which gives the most reliable results.

← Scroll to see the full table →

ArrangementTransducer placementAccuracyBest for
Direct (opposite faces)Transmitter and receiver on opposite faces, facing each otherHighest: straight pulse path, clearly defined distanceColumns, beams and walls accessible from two sides
Semi-direct (adjacent faces)Transducers on two perpendicular faces; L is the diagonalMediumColumn and beam corners accessible from two adjacent faces
Indirect (same face)Both transducers on the same surfaceLowest: mainly reflects near-surface qualitySlabs, walls accessible from one side only, and crack depth assessment

Couplant (coupling gel) must always be applied to the transducers for good pulse transmission, and the actual distance between transducers must be measured, since it directly affects the velocity value.

SPN team on scaffolding placing UPV transducers on the concrete slab soffit and beams above the ceiling inside a building
The SPN team on scaffolding to reach the concrete slab soffit and beams above the ceiling — site access determines whether the transducers are placed in the direct, semi-direct or indirect arrangement

The figure below shows the three UPV transducer arrangements as 2D sections, with a worked velocity calculation and a caution about reinforcement, followed by a 3D model of a column tested on a grid of points to produce a velocity map — the basis for reading the results in the next section.

2D model of the three UPV transducer arrangements, a worked pulse velocity calculation and a caution about reinforcement The top row shows sections of three concrete members. In the direct arrangement, transmitter T and receiver R are on opposite faces of a column; the pulse travels straight and L is the section width, giving the most accurate result. In the semi-direct arrangement, the transducers are on two perpendicular faces and L is the diagonal. In the indirect arrangement, both transducers are on the same face; the pulse travels shallowly beneath the surface and accuracy is lowest. Couplant gel must be applied to the transducers in every case. The bottom left shows a worked example: a path length of 250 millimeters and a travel time of 62.5 microseconds give a velocity of 4.0 kilometers per second. The bottom right is a caution: if the pulse path runs parallel to reinforcement, the reading will be falsely high, so choose a path that avoids or crosses the bars; moist concrete also gives higher velocity. 2D section — transducer arrangements Not to scale 1 Direct (opposite) Most accurate · columns, beams 2 Semi-direct (corner) Two adjacent faces accessible 3 Indirect (same face) Least accurate · shallow cracks T R L = section width T R L = diagonal √(a² + b²) a b T R L = probe spacing Shallow path: mainly surface quality Couplant at every probe placement · measure actual L, it directly affects velocity Unit reads time, then V = L ÷ T 62.5 µs L = 250 mm (measured) T = 62.5 µs V = 250 ÷ 62.5 = 4.0 mm/µs = 4.0 km/s Caution: rebar carries pulse faster ✓ ✕ Path parallel to bars → velocity too high ✓ Locate bars first, then avoid or cross them Moist concrete also reads faster
2D model — Direct places the transducers on opposite faces and gives the most accurate result; Semi-direct is used when two perpendicular faces are accessible, with L as the diagonal; Indirect places the transducers on the same face and has the lowest accuracy. Worked example: 250 mm path, 62.5 µs gives 4.0 km/s. The bottom-right panel warns that a pulse path parallel to reinforcement gives a falsely high value. (Image: SPN Soil Engineering)
3D model of a concrete column tested by direct-transmission UPV on a grid of points, and a pulse velocity map identifying an area of poor-quality concrete 3D view of a concrete column on a floor, shown semi-transparent. The front face has a grid of test points in 4 columns and 6 rows, with transmitter T on the front face and receiver R on the opposite back face. In the upper part the pulse travels straight through dense concrete; in the lower part near the column base there is internal honeycombing, so the pulse must detour, increasing travel time and reducing velocity. The right panel is a pulse velocity map of the same grid: the upper part is approximately 4.2 to 4.5 kilometers per second, while the lower middle drops to 2.9 to 3.3, with a color bar of general reference criteria following the Whitehurst guideline and a note to look at uniformity; points with low values should be checked further by other methods or by coring. 3D model Not to scale T Test grid on face Dot color = velocity range Direct path, dense Short time → high velocity Transmitter T (front) Receiver R, back face Internal honeycomb Detour → long time Low velocity Pulse velocity map (km/s) 4.4 4.5 4.4 4.3 4.3 4.4 4.5 4.4 4.4 4.3 4.4 4.2 4.2 3.9 4.0 4.3 3.8 2.9 3.1 4.0 4.0 3.3 3.6 4.1 Same grid as column (top → bottom) General reference (Whitehurst guideline) > 4.5 Excellent 3.5–4.5 Good 3.0–3.5 Fair/doubtful 2.0–3.0 Poor < 2.0 Very poor Look at “uniformity,” not one value Red box: low cluster near the base → Possible void: verify by other methods/cores Values shown are illustrative only Strength conversion needs core calibration
3D model — direct-transmission readings are taken point by point on a grid across the column face. In the upper part the pulse passes through dense concrete; in the lower part it must detour around honeycombing, so the velocity drops as a cluster. The map on the right is used to judge uniformity against general reference criteria. Values are hypothetical, for illustration only. (Image: SPN Soil Engineering)

What Are the Steps of a Field UPV Test?

A field UPV test has 6 main steps. Most errors come from surface preparation and distance measurement, not from the instrument itself.

  1. Define locations and the test grid — the engineer selects the members and grid spacing to suit the objective, e.g. checking uniformity of the whole member or targeting a suspect area.
  2. Locate reinforcement first — use a cover meter or GPR concrete scanning to avoid test paths parallel to the bars.
  3. Prepare the surface down to sound concrete — remove plaster or paint only at the transducer locations, then grind smooth.
  4. Check the instrument against the reference bar — zero the instrument with the calibration bar supplied with it before starting.
  5. Apply couplant, press the transducers firmly and read the time — repeat readings until stable, and measure the actual distance L at that point with a measuring tool.
  6. Calculate velocity and map the results — record the raw time, distance and velocity at every point for point-by-point comparison within the same structure.

How Do You Read UPV Velocity Values?

Pulse velocity indicates concrete quality in relative terms. The table below gives widely used general reference criteria (based on the Whitehurst guideline for normal-weight concrete), which should be adjusted for the mix and interpreted by an engineer.

← Scroll to see the full table →

Pulse velocity (km/s)General concrete quality
Above 4.5Excellent
3.5 – 4.5Good
3.0 – 3.5Fair/doubtful
2.0 – 3.0Poor
Below 2.0Very poor

The key principle is that UPV is best for judging "uniformity" rather than single values. If the velocity drops significantly at any location within the same structure, that area may contain voids, cracks or poor-quality concrete and should be checked further by other methods. Details on interpreting low results and confirmation steps are in the article What Does a Low UPV Test Result Mean?

SPN engineer reading pulse velocity on the UPV instrument while testing floor concrete quality inside a building
An SPN engineer reading the pulse travel time from the UPV instrument on site — the reading is divided into the transducer spacing to calculate pulse velocity

What Can UPV Be Used to Check?

UPV can be used both to assess overall quality and to locate specific defects. Common applications include:

  • Checking concrete uniformity throughout a structure, point by point
  • Detecting voids or honeycombing inside the concrete
  • Estimating the depth of surface-opening cracks
  • Monitoring deterioration from chemical attack, aging or fire damage
  • Combining with the rebound hammer (SonReb method) for a more reliable compressive strength estimate
SPN team performing UPV testing of slab soffit and beam concrete at an opened ceiling inside a building
Checking the quality of slab soffit and beam concrete inside a building — UPV is well suited to point-by-point uniformity comparison across a whole structure

What Are the Limitations of the UPV Test?

UPV is accurate in relative terms, but the velocity changes with several site factors not directly related to concrete quality. The table below summarizes common factors and how to deal with them.

← Scroll to see the full table →

FactorEffect on the measured valueHow to deal with it
Pulse path parallel to reinforcementFalsely high velocity, because steel transmits the pulse faster than concreteLocate the bars first, then test along a path that avoids or crosses them
Moist or saturated concreteHigher velocity than in dry conditionCompare only points with similar moisture conditions
Rough surface or insufficient couplantFalsely long transit time, velocity lower than actualGrind the surface smooth, reapply couplant and retest
Error in measuring LVelocity in error in proportion to the distance errorMeasure the actual distance at every point; do not use drawing dimensions
Different concrete mixes and agesDifferent baseline velocitiesCompare within the same concrete batch, not across structures

As with the rebound hammer, converting UPV to compressive strength requires calibration against core tests of the same concrete. Compressive strength should not be concluded from pulse velocity alone.

Which Tests Should UPV Be Combined With?

UPV gives an indication, not a verdict, so engineers usually combine it with other methods depending on the question to be answered.

  • To estimate compressive strength — pair it with the rebound hammer, then calibrate with core tests only where necessary.
  • One-sided access, to find voids or measure thickness — supplement with Impact Echo.

SPN provides non-destructive structural evaluation (NDE), with reports certified by a civil engineer licensed by the Council of Engineers Thailand (COE), complete with signature and registration number.