Preload Force vs. Torque: Why You Are Measuring the Wrong Parameter - And How to Do It Right

You monitor torque. Your tightening tool reports "OK", the process is running. Everything under control - or is it?

Here is the uncomfortable truth: Only around 10% of the applied tightening torque actually generates clamping force (preload) in the bolt connection - the remaining ~90% is lost to friction. Roughly speaking, about 50% of this friction occurs under the bolt head and about 40% in the threads - as described in standard technical documentation on bolting technology.

This means: if you only monitor torque, you are not controlling the holding force of your bolt connection. You are monitoring the input - not the result.

For engineers in automotive production, aerospace, or mechanical engineering, this is far from an academic detail. It is the core difference between a truly reliable bolt connection and one that only appears to be.

The physics of a bolt connection - in a nutshell

What is preload (clamping force)?

Preload is the actual holding force of a bolt connection. It is generated when the bolt is stretched axially during tightening - much like a tensioned spring. This tensile force in the shank clamps the jointed parts together and creates the clamping force that is critical for the function of the connection: it keeps components firmly seated without play, secures sealing surfaces, and decisively determines fatigue strength and service life.

What is torque - and why is it not enough?

Torque is force multiplied by lever arm - the parameter you set on your torque tool. It is easy to measure and straightforward to implement technically. That is precisely why torque monitoring has become the standard control variable in tightening technology and torque analysis.

The problem: Torque is still the most commonly used control variable in bolting technology, but the resulting preload is heavily influenced by fluctuating friction coefficients and by the torque scatter of the tightening device. Friction is therefore the central - and largest - unknown.

Where does the energy go?

The distribution is clear: Only about ten percent of the applied tightening torque is converted into clamping force. The remaining tightening energy is consumed by friction losses in the bolt connection: as a rule of thumb, 40% of the torque is lost to friction in the threads, 50% to friction under the bolt head.

This is why any serious bolt connection analysis has to look beyond torque monitoring alone and consider friction loss as a dominant influence.

The friction coefficient - your biggest unknown

What influences the friction coefficient μ? More factors than most engineers account for in their torque monitoring strategy:

• Surface condition: Roughness, oxides, contamination
• Coating: Zinc-plated, phosphated, uncoated - every coating has a different μ value
• Lubricant: Type, quantity, and uniformity of application
• Temperature: Elevated temperatures change viscosity and friction behavior
• Material pairing: Steel on steel behaves fundamentally differently than stainless steel on aluminum - some combinations tend to galling and cold welding

Each of these factors directly affects friction loss and therefore the correlation between tightening torque and the actual preload torque and preload measurement in your bolt connection.

Why identical torque ≠ identical clamping force

Imagine two identical bolts tightened with the same tightening torque - one freshly coated, the other slightly corroded. Same tool, same setting, same operator.

The result in the bolt connection? Potentially very different.

Fluctuating friction conditions mean that even with very high torque repeatability, variations in the resulting preload of 50% or more are possible. This is not theory - it is measurable reality in every serial production process that relies only on torque verification.

The conclusion many design engineers draw: over-dimensioning. Bolts are sized so that they still hold with minimal clamping force and are not overloaded at maximum clamping force. Safety margins are planned rather than understood.

This approach works - until the connection becomes safety-critical. Until an auditor requests documented evidence of process capability. Until a field failure occurs where the torque protocol still shows everything as "OK".

If you are interested in the normative requirements for safety-critical bolt connections, read our article on Class A bolted joints according to VDI/VDE 2862 - there you will learn why documented tightening torque alone is not sufficient in critical cases.

The solution: combined torque-angle analysis

Why angle provides the missing information

The angle of rotation is the direct geometric equivalent of bolt elongation: Axial force is proportional to torque - elongation is proportional to angle. Once you know how far the bolt turns after head seating, you can directly infer the actual elongation - and therefore the clamping force - from this angle measurement.

This is the decisive advantage: angle is largely independent of friction. While tightening torque is heavily influenced by fluctuations in μ, the physical relationship between angle and elongation remains stable.

Torque angle analysis closes exactly this information gap.

The torque-angle curve as the "fingerprint" of the connection

Only the combination of torque and angle delivers the complete picture. The resulting torque-angle curve (also called tightening curve) shows in a single diagram:

1. Run-down phase: The bolt turns with negligible resistance
2. Head seating / snug point: Torque begins to rise - force build-up starts
3. Elastic region: Linear increase - torque and angle correlate directly with preload
4. Yield point: The curve flattens - the bolt begins to deform plastically
5. Plastic region / fracture zone: Curve drops - overload

Deviations in this curve - flat progressions, unexpected jumps, overly steep rises - are clear indicators of process issues: incorrect friction values, thread defects, missing lubricant, or incorrect component combinations.

How to interpret this curve in detail is the focus of the follow-up article in this series: Torque-angle analysis: How to read the "fingerprint" of your bolt connection (coming soon). This is where angle analysis and angle monitoring become practical tools for engineers.

Tightening methods at a glance: what really matters

Not every tightening method is suitable for every joint. The following overview shows how the three common methods differ in accuracy and friction dependence and how they influence torque monitoring and bolt connection analysis:

Torque-controlled tightening

By far the most widespread method - simple and cost-effective. The only control variable is torque. Due to the strong dependence on friction, bolts are typically designed to only 60 to 70% of yield strength to avoid the risk of fracture in the event of higher friction. The result: a wide scatter in the actual clamping force and systematic over-dimensioning as a standard strategy.

This is the classic domain of basic torque monitoring, but it provides only a limited view of the real state of the bolt connection.

Angle-controlled tightening

Here, a defined pre-torque is applied first, followed by a specified angle of rotation. Once the contact surfaces have settled and the bolt is working in a stable region, an additional angle of rotation correlates more reliably with additional elongation than torque alone. The scatter of preload decreases significantly. This method is mainly used for safety-critical bolt connections where higher reproducibility is required and where torque angle analysis delivers clear advantages.

Yield-controlled tightening (YCA)

The most precise method: the bolt is tightened right up to the yield point - identified by a drop in the gradient dM/dφ in the torque-angle curve. The advantage of yield-controlled tightening compared to torque-controlled tightening is that, for the same scatter of the thread friction coefficient, the scatter of the installed preload is lower. The bolt is utilized to the maximum - which, in lightweight design applications (automotive, aerospace), allows the use of smaller and lighter bolts.

The prerequisite: highly accurate torque monitoring and angle gauge technology that can detect the curve gradient in real time.

For more in-depth information on process control and capability evidence, we recommend our article on process capability studies (PFU) according to VDI/VDE 2645-3.

QUANTEC MCS® - the tool for complete bolt connection analysis

This is precisely where QUANTEC MCS® from GWK comes in. It was developed to record the complete torque-angle curve in real time - and thus provide what pure torque monitoring cannot: the fingerprint of your bolt connection.

If you are looking for advanced torque analysis and torque verification in development, quality assurance, or production, QUANTEC MCS is designed for you.

Angle measurement without fixed reference

The core of QUANTEC MCS® is its reference-free angle measurement - angle is recorded without a fixed external reference point. The tool can be positioned freely, with no need for prior zeroing. Errors due to incorrect referencing are structurally eliminated. The resolution is 0.1° - fine enough for yield point analysis even on demanding small fasteners.

This turns QUANTEC MCS into a powerful angle gauge and angle monitoring system for modern torque bolt applications.

Key technical data at a glance

• Measurement accuracy: ±1% between 10 and 100% of the nominal range
• Construction: Robust aluminum-titanium design (no carbon tubes as used by many competitors) for long-term accuracy and durability in production environments
• Data transmission: Wireless WLAN transfer directly to the evaluation software
• Software compatibility: QuanLabPro, Ceus, and QS-Torque
• Additional channel: 16-bit measurement channel for external preload sensors (e.g. piezo) for direct clamping force and preload measurement

From analysis to process optimization

QUANTEC MCS® is not just a test instrument - it is an analysis tool. With it, you can:

• Investigate existing tightening processes for the real reproducibility of preload and tightening torque
• Validate and optimize tightening methods (torque-controlled, angle-controlled, or yield-controlled)
• Perform friction value analyses for different lubricants, coatings, and material pairings to minimize friction loss
• Detect curve deviations that indicate process problems - before they lead to failures in the field
• Establish a solid data basis for process capability studies (PFU) according to VDI/VDE 2645-3 - as we discuss in detail in our article on Cmk and Cpk in bolting technology

QUANTEC MCS® is also available flexibly through GWK ToolRent® - with weekly, monthly, or annual rental options and worldwide shipping. Ideal for one-off projects, audits, or bridging the period until a capital investment decision is made.

Conclusion: if you only measure torque, you control the input - not the result

Torque is an input parameter. Preload is the result. If you only monitor the former, you have no reliable statement about whether the bolt connection is actually secured as assumed.

For standard joints, pure torque monitoring may be sufficient. For safety-critical connections in automotive, aerospace, rail, and mechanical engineering - for any bolted joint where failure has consequences - you need more. You need the complete view of the bolt connection.

Torque-angle analysis provides this view. QUANTEC MCS® makes it measurable, traceable, and repeatable - from development through series production.

Accuracy by GWK.