A technician places a torque wrench on an already-tightened fastener, applies the target value - no click, no movement. Result: "Pass." But that conclusion doesn't hold up physically. What was actually measured is not the assembly torque from the original installation; it's an artifact of static friction and settling forces. Failing to understand this distinction risks incorrect approvals and gaps in your audit trail.
This guide is written for quality assurance and audit professionals who want to understand what is physically happening during a post-assembly fastener inspection - and how to do it right, methodically.
The Core Problem: Why Simply Re-Tightening Tells You Nothing Reliable
When a fastener is initially tightened, kinetic friction is at work: the fastener rotates, and the thread and bearing surface slide against each other. The applied tightening torque generates the desired clamp load.
The moment the fastener stops moving, the friction physics change. Static friction builds between the fastener head and the mating surface - and static friction is inherently higher than kinetic friction. On top of that, under high surface pressure, molecules from the fastener and the substrate can bond together; that bond must be broken before the fastener will move at all. Breakaway torque therefore depends solely on static friction and says nothing whatsoever about the original assembly.
When a technician then re-tightens to the target value with a torque wrench, they are not measuring the original assembly torque. They first overcome the breakaway torque - and only after that does the fastener continue to turn. The original tightening occurred under kinetic friction; re-tightening first requires overcoming static friction. The result is therefore not comparable to the assembly condition.
There is also the so-called re-tightening factor: VDI guideline 2230 defines the re-tightening torque as the torque required to continue rotating a fastener after the tightening process is complete. It differs from the target assembly torque by the re-tightening factor, which - depending on the type of tool, friction conditions, and elasticity ratios - can range from 0.85 to 1.30. Tolerances of this magnitude make it impossible to determine whether a fastener was installed within tight process limits.
Three Terms, Clearly Defined
Before describing the inspection method, three terms must be clearly distinguished from one another. In practice, they are frequently confused.
Breakaway Torque (M_LH)
Breakaway torque is the torque required to first release a stationary, preloaded fastener from static friction - regardless of whether the direction is tightening or loosening. The breakaway torque when loosening M_LH or continuing to turn M_WH is the torque required after the tightening process is complete to overcome the static friction of the fastening element. It is not a direct measure of clamp load and is therefore only conditionally suitable for quality assurance purposes.
Prevailing Torque (M_WH)
Prevailing torque is the torque measured after the breakaway torque has been overcome and the fastener is once again in motion - that is, in the transition from static to kinetic friction. The prevailing torque M_WH is the torque required after the tightening process is complete to overcome the static friction of the fastening element. It correlates significantly better with the original assembly torque than breakaway torque does, because it reflects the condition of the joint in the kinetic friction regime.
The most widely used method in practice for determining assembly torque is measuring the so-called prevailing torque. To do this, the inspector places an indicating torque wrench on the already-installed fastener and rotates it a small amount further. The value obtained is recorded and evaluated statistically as a measure of the installed torque accuracy.
Residual Torque / Check Torque
The term residual torque (also: check torque) is often used interchangeably with prevailing torque in practice, but more precisely describes the torque remaining in the joint after the assembly process is complete - accounting for settling and relaxation effects. For the process capability study per VDI/VDE 2645-3, prevailing torque is the normatively defined measurement variable.
| Begriff | Definition | Eignung für QS-Prüfung |
|---|---|---|
| Losbrechmoment (M_LH) | Moment zum Überwinden der Haftreibung aus dem Stillstand | Bedingt – stark von Haftreibung abhängig, kein direkter Bezug zur Vorspannkraft |
| Weiterdrehmoment (M_WH) | Moment im Gleitreibungsbereich nach dem Losbrechen | Hoch – normativ definiert in VDI/VDE 2645-3, statistisch auswertbar |
| Restmoment / Prüfmoment | In der Verbindung verbleibendes Drehmoment nach Montage und Setzen | Hoch – Praxisbegriff, oft synonym mit Weiterdrehmoment verwendet |
The Normative Framework: VDI/VDE 2645-3
VDI/VDE guideline 2645, Part 3 describes procedures for process capability studies (PCS) for prevailing torques of bolted joints with clamp load. The objective of a process capability study for fastening processes is to evaluate and document the quality capability of a bolting process under production conditions.
This includes identifying systematic influences for targeted improvement of the bolting process, assessing the effect of process improvement measures, evaluating tolerances for the process capability study, and establishing the control limits for quality control charts.
A key distinction from a machine capability study (MCS): unlike a machine capability study, a process capability study accounts not only for machine influence but also for the influence categories of people, material, method, and environment. The PCS therefore evaluates the real production process - not just the tool in a lab setting.
One further critical point: repeated measurement of the prevailing torque characteristic on the same fastener - as is common practice for quantifying measurement process suitability - is not possible. Every loosening or further rotation irreversibly changes the state of the test object. This demands careful planning of the sampling strategy.
Step by Step: The Correct Prevailing Torque Inspection
Before the inspection, define: Which bolted joints will be inspected? What sample size will be used (VDI/VDE 2645-3 typically recommends n ≥ 25 measurements per bolted joint for a PFU)? What tolerance limits apply? Determine whether the inspection will be conducted as a sample during ongoing production or as a dedicated PFU campaign. Ensure that the inspection tool is calibrated with DAkkS-traceable calibration and that the calibration is current.
Select an inspection tool designed for prevailing torque measurement — not a simple torque wrench with a drag pointer. The tool must be able to resolve the torque-angle curve in order to cleanly distinguish between breakaway torque and prevailing torque. Set the measurement range so that the expected reading falls between 10 and 100% of the nominal range — measurement accuracy is only within specification in this range.
Place the inspection tool on the already-tightened bolt. Continue turning in the tightening direction with a smooth, slow rotational motion. The tool captures the torque-angle curve: first the rise to breakaway torque, then the drop and stabilization in the sliding friction range — this is the prevailing torque. Record the measured value and the corresponding angle of rotation. Important: Each bolt may only be measured once, as the condition of the joint is irreversibly altered by the measurement.
Enter the measured values into your analysis software. Calculate the mean, standard deviation, and the process capability index Cpk. According to VDI/VDE 2645-3, a bolting process is considered capable when Cpk ≥ 1.67 (for safety-critical joints) or Cpk ≥ 1.33 is achieved. If measured values fall outside the tolerance limits, a root cause analysis must be initiated — tool drift, changes in friction coefficients, or process parameters are typical causes.
Document the following for each inspection: bolted joint (identification), inspection date, inspector, inspection tool used (serial number, calibration certificate), measured values (prevailing torque, angle of rotation), statistical parameters, and evaluation result. Traceability of the measurement to national standards must be demonstrated via the inspection tool's calibration certificate. Archive the raw data in a way that ensures it can be reproduced and presented at any time during an audit.
Common Sources of Error - and How to Avoid Them
1. Documenting breakaway torque as the inspection result The most common mistake: the tool displays the peak value at breakaway, and that value is recorded as the prevailing torque. Breakaway torque and prevailing torque are physically distinct quantities. Only a tool that records the complete torque-angle curve allows the two values to be correctly separated.
2. Inspecting in the loosening direction Some inspectors rotate the fastener in the loosening direction to determine the "loosening torque." The loosening torque is systematically lower than the prevailing torque in the tightening direction, because the self-locking effect of the thread is reduced when loosening. Breakaway torque is not equal in both directions - tightening increases self-locking, while loosening reduces it. For a PCS per VDI/VDE 2645-3, the tightening direction is always the reference.
3. Taking multiple measurements on the same fastener Measuring the same fastener twice to gain "confidence" corrupts the result. Every loosening or further rotation irreversibly changes the state of the test object. Each fastener may only be measured once.
4. Measuring outside the calibrated range If an inspection tool with a nominal range of 100 Nm is used on a fastener with an expected prevailing torque of 8 Nm, the measurement falls below 10% of the nominal range. In this region, measurement accuracy is not within specification - the result is meaningless.
5. Non-traceable calibration In an audit, traceability of the measurement to national standards must be demonstrated without gaps. A calibration certificate without accredited laboratory accreditation or without a stated measurement uncertainty does not meet this requirement.
6. Failing to distinguish between MCS and PCS The most pervasive error in quality assurance: machine capability studies and process capability studies are treated as equivalent or confused with each other. They are not - they answer fundamentally different questions. An MCS conducted in a lab demonstrates tool capability; it does not replace a PCS under production conditions.
The Right Tool: Q-CHECK® for QA and Audit
AI-generated imageThe inspection task described here requires a tool specifically designed for prevailing torque measurement and the PCS per VDI/VDE 2645-3 - not a standard torque wrench, not a calibration device.
The Q-CHECK® from GWK was developed precisely for this purpose: as a QA and audit tool for prevailing torque measurement on already-tightened joints.
Key technical specifications at a glance:
- Measurement range: 3-1,000 Nm - covers the vast majority of industrial fastening applications
- Measurement accuracy: ±1% between 10 and 100% of the nominal range - specification-compliant for the PCS
- Storage: 2 GB internal memory for up to 1,000 fastening points - all measurement data remains on the device until transferred
- Traceability: DAkkS-accredited calibration - complete audit trail
- Construction: Robust aluminum-titanium design for use in production environments
The Q-CHECK® captures the complete torque-angle curve. This allows breakaway torque and prevailing torque to be cleanly separated - the decisive advantage over a simple drag-pointer wrench. Measurement data is stored directly on the device and can be exported for statistical evaluation per VDI/VDE 2645-3.
Want to know whether your current inspection method for prevailing torque measurement is compliant with the relevant standards? Our experts will analyze your bolting process and show you exactly what data you need for a robust PFU in accordance with VDI/VDE 2645-3.
Request a free screw process analysisInteractive Decision Guide: Which Inspection Method Fits Your Application?
Documentation: What Counts in an Audit
A prevailing torque inspection is only as reliable as its documentation. The following records must be available for review in an audit at any time:
| Record | Requirement |
|---|---|
| Inspection plan | Fastening points, sample size, tolerance limits, inspection interval |
| Inspection tool calibration | Accredited calibration certificate, current, with stated measurement uncertainty |
| Measurement data (raw data) | Prevailing torque, rotation angle, timestamp, inspector, fastening point |
| Statistical evaluation | Mean, standard deviation, Cpk - referenced to VDI/VDE 2645-3 |
| Actions for deviations | Root cause analysis, corrective action, effectiveness verification |
Traceability of the measurement to national standards is not optional - it is mandatory. Only when the inspection tool has been calibrated by an accredited laboratory is the measurement uncertainty quantified - and with it, the statement about process capability made defensible.
Conclusion: Method Before Tool
Inspecting an already-tightened fastener is not a trivial re-tightening exercise. It requires an understanding of the underlying physics - static friction, breakaway torque, prevailing torque - and a method that cleanly separates these quantities.
VDI/VDE 2645, Part 3 defines prevailing torque measurement as the normative procedure for process capability studies of bolted joints with clamp load. Using this standard as the foundation provides a defensible basis for QA decisions and audit documentation.
The right tool - one that captures the complete torque-angle curve, is calibrated with full accredited traceability, and stores data in a tamper-proof, revision-safe manner - is not optional. It is a prerequisite for a standard-compliant result.
Accuracy by GWK.
