Ultrasonic Bolt Tension Meter Guide

Measurement model

Time of flight becomes useful through calibration.

An ultrasonic bolt tension meter does not read force directly from an echo. It measures an acoustic quantity—usually a change in time of flight—and applies a calibrated relationship for the bolt material and configuration. The output can then be expressed as elongation, stress or axial force, depending on the method and inputs.

A primary research paper on ultrasonic bolt preload without an unloaded reference describes the need for precise time-of-flight measurement and investigates how ultrasonic velocity changes with preload. It also demonstrates why the measurement model, wave mode and test population matter; a technique developed in a study should not be assumed to exist in every portable meter.

Bolt stretch means elongation, not tightening torque.

“Bolt stretch measurement” commonly describes the change in bolt length under load. In an ultrasonic workflow, the measured time of flight and the material calibration must be distinguished from a mechanical length measurement. A stretch value alone does not specify axial force without the relevant geometry and material relationship.

For a measurement engagement rather than an instrument purchase, see the ultrasonic bolt load verification service. Availability, procedure, personnel, report format and calibration traceability are settled for the proposed job; the product specification is not a substitute for that review.

Keep five quantities separate

• Torque: the rotational input applied during tightening.
• Preload or axial force: tensile force in the bolt before external service load.
• Elongation: change in effective bolt length under load.
• Ultrasonic time of flight: measured pulse travel time through the selected acoustic path.
• Stress estimate: a calculated result that depends on material and geometric inputs.

Method selection

Compare what is controlled and what is inferred.

Torque, hydraulic pressure and ultrasound observe different parts of the tightening process. The methods can be complementary. Selection depends on joint design, required uncertainty, access, installation sequence, remeasurement needs and the approved engineering procedure.

ISO 16047 specifies conditions for torque/clamp-force testing of relevant threaded fasteners. That standardised test relationship is different from measuring the actual preload of an installed wind-turbine bolt with ultrasound.

If a tightening method has not been chosen yet, start one step earlier: bolt torque chart and preload calculator puts numbers on that spread; torque vs tension across the recognised pretensioning methods compares what each one controls and how conformity is evidenced.

If the measurement is needed but the instrument is not, it can be bought as work rather than equipment: what an ultrasonic bolt measurement engagement produces sets out the output, the inputs it needs and what is settled in scoping.

Measurement controls

Record the inputs that can move the result.

Repeatability depends on keeping the acoustic and mechanical state traceable. An apparently stable number can still be misleading if the reference changed, a different echo was selected or the material coefficient came from an unmatched bolt population.

1. 01Identify the bolt

   Record manufacturer or batch where available, material, property class, diameter, effective length and end geometry.

2. 02Prepare the path

   Confirm parallel or approved contact faces, surface condition, probe location, couplant and a stable back-wall echo.

3. 03Establish calibration

   Document the unloaded reference or alternate qualified method, material coefficient, load reference and calibration range.

4. 04Control conditions

   Record bolt and ambient temperature, probe, coupling, instrument setup and any timing between tightening and reading.

5. 05Apply the sequence

   Follow the approved tightening pattern, target, tolerance, hold points and remeasurement rule for the joint.

6. 06Retain the record

   Store raw readings, calculated force, tolerance result, tool state, operator, timestamp and exceptions.

Common invalidation risks

• A material coefficient copied from a different alloy, heat treatment or bolt batch
• Temperature change between baseline and loaded measurements without a controlled correction
• Curved, rough, coated or nonparallel contact surfaces that make echo selection unstable
• A crack-like indication, geometry echo or mode conversion obscuring the selected back-wall signal
• Plastic deformation or operating load outside the calibration model
• Reporting only the final force value without raw reading, setup and tolerance context

Wind-turbine configuration

Do not merge preload and integrity into one result.

A preload result answers a load question; an NDT result answers a defined discontinuity question. The same ultrasonic platform may contribute to both only when the procedure establishes suitable acoustic paths, calibration and interpretation for each task.

The 2024 University of Strathclyde wind-turbine bolt PAUT study investigated stress measurement and defect detection together. The authors calibrated an acoustoelastic constant and reported that the value is material-specific; their M42 blind test also retained uncertainty and further-development limitations. It is method evidence, not a universal installed-bolt accuracy specification.

NASA’s official Fastener Design Manual identifies friction at threads and bearing surfaces, coatings and lubricants among the variables that complicate torque-to-preload selection. This supports a procurement distinction: controlling tightening torque is not identical to measuring axial force.

Configuration inputs for an engineering review

• Bolt drawing, material or batch data, target preload and allowable tolerance
• Joint location, tightening sequence, existing tooling and available reaction/access space
• End-face condition, effective ultrasonic path and whether an unloaded reference can be captured
• Temperature range, coating, lubricant, reuse policy and expected measurement interval
• Required raw data, audit fields, controller interface and acceptance authority

If the project also requires crack-like indication detection, compare the PAUT, FMC and TFM guide and define a separate inspection volume and acceptance procedure.

Verification vs monitoring

A measurement is a moment; monitoring is a series.

Ultrasonic axial-force measurement produces a dated record of one tightening or one re-check, not a standing view of the joint. The distinction matters at procurement, because a specification that asks for "bolt tension monitoring" and a specification that asks for verified preload at installation are satisfied by different things.

What the retained record does tell you

• The axial force reached at that tightening, against the configured target and tolerance
• The bolt parameters, calibration state and conditions the value was obtained under
• An audit trail that a later re-check can be compared against, bolt by bolt

What it does not tell you

• What the preload is now, if the joint has seen service since the record was written
• Whether relaxation, embedment or settling moved the load after tightening
• Whether a discontinuity has developed in the bolt — that is a separate inspection question with its own procedure

Continuous in-service load monitoring is a different instrument class with its own sensors, power, data path and calibration drift behaviour. WindInspectTech's published bolt-integrity documentation covers measurement and record retention at tightening; it does not document a permanently installed load-monitoring channel. Where a programme genuinely needs the standing value rather than the installation record, that requirement should be stated separately in the enquiry so it is not assumed to be covered.

For the equipment and its documented conditions, see the WIT-BW10 specifications. For the separate question of finding crack-like indications in a bolt, see the PAUT, FMC and TFM guide.

Procurement

Read the accuracy figure before the price.

Two meters quoting "3% accuracy" can mean different things. Full scale, of reading, at a stated temperature, on a qualified bolt — each qualifier changes what the number covers, and a catalogue rarely says which it is. A buyer comparing distributor listings is comparing numbers with different definitions.

Six questions that separate a measurement from a product

1. Accuracy basis. Is the figure full scale or of reading, and at what load range? WIT-BW10 documents 3% full scale; at loads well below full scale the absolute uncertainty is the same, so the relative uncertainty is larger. That is arithmetic, not a defect — but it has to be known before the tolerance is set.
2. Qualified bolt range. Which sizes, lengths and materials has the transducer path actually been qualified on? A ceiling such as ≤M64 is a limit, not a guarantee of equal performance across the range.
3. Reference state. Does the method require an unloaded baseline reading, and if a reference-free mode is claimed, what uncertainty is stated for it? Research methods exist; not every instrument implements one, and the uncertainty is the whole point.
4. Temperature handling. Is compensation automatic from a measured value, entered manually, or absent? The bolt on a nacelle at 06:00 and at 14:00 is not at the same temperature.
5. Record and export. What is stored per tightening — target, achieved value, bolt parameters, calibration state, time — and in what format can it leave the instrument? A record that cannot be exported cannot be audited later.
6. Coupling and end-face requirements. What surface condition does the probe need, and who prepares it? On a flange with a hundred bolts this is the labour cost, not the meter.

When ultrasonic is the wrong purchase

• The bolts have no accessible, prepared end face and the programme will not fund preparing one — the measurement path does not exist.
• The joint is qualified on torque with controlled fasteners and the specification asks for torque records, not preload — the meter would produce a number the acceptance criteria cannot use.
• The requirement is actually continuous in-service monitoring — see the previous section; that is a different instrument class.
• The bolt population is small and one-off, so the per-bolt calibration effort exceeds the value of the record.

The documented conditions for the equipment this site publishes are on the WIT-BW10 page. Put the six questions above to that page too; where it does not answer one, that is the question to raise in the enquiry rather than assume.