An Engineer’s Guide To Bolt Preload and Clamping Force

Tighten a bolt and something slightly counterintuitive happens: it stretches. Not by much, but that tiny bit of stretch is what makes the whole joint work. The bolt behaves like a very stiff spring and the tension locked into it pulls the clamped parts tightly together. That tension is called preload and it's the single biggest factor in whether a bolted joint holds or fails. The compression it creates in the parts being clamped is the clamping force. Get preload right and a joint stays tight for years. Get it wrong and you've compromised bolted joint integrity, which sits behind a surprising share of mechanical failures.

This guide covers what preload and clamping force actually are, how they relate to one another, what goes wrong when they're off and how to keep them under control. Accu's engineers have weighed in throughout, so whether you're torquing down a critical joint or just want to understand what's happening under the spanner, there's something here for you.

 

Contents:

• What are Bolt Preload and Clamping Force and How Do They Work?

• What Happens When You Get It Wrong

• Torque and Preload in Bolted Joints

• The Factors That Affect Preload and Clamping Force

• Methods of Controlling Preload and Clamping Force

• Common Mistakes and How to Avoid Them

• Key Takeaways

• FAQs

What are Bolt Preload and Clamping Force and How Do They Work?

 

When you tighten a bolt, you set two forces working against each other, equal in size but opposite in direction. The first is preload, the tension that builds inside the bolt as it's tightened. The second is clamping force, the compression that same action squeezes into the joint.

Preload comes from the bolt stretching. As you turn the nut, the bolt is pulled tight and stretches by a small amount, and like any stretched spring it fights to pull back to its original length. That resistance, the pull it exerts trying to spring back, is the preload.

That pull has to act against something. The nut and the bolt head trap the joint between them, so as the bolt tries to contract, it squeezes the clamped parts together. That squeeze is the clamping force.

Preload and clamping force are interlinked, as you can’t have one without the other. They’re also normally equal in magnitude. If there are 10 kilonewtons (kN) of preload, the clamping force will also be 10 kN.

What Happens When You Get It Wrong

 

There are two main routes to failure when it comes to preload in bolts: undertightening and overtightening. It’s fairly self-explanatory from the names themselves how you can arrive in either of these states, but what happens when you get preload wrong on the bolt and it’s either undertightened or overtightened?

Start with undertightening. A bolt that's too loose can't generate the clamping force the joint needs to stay put, so the joint slips. Once it slips, shear load lands directly on the bolt shank instead of being carried by friction between the clamped faces, and the risk of the bolt shearing climbs. Fatigue life takes a heavy hit too, because a loose bolt sees the full swing of every load cycle rather than a gentle fluctuation.

That's not the end of it. In sealed joints, too little clamping force means leaks. And as anyone who's watched a Junker test will tell you, an undertightened joint under cyclic vibration will happily walk its own nut loose and undo itself.

Overtightening brings a different set of problems. Some are immediate and obvious: stripped threads, a sheared bolt or clamping faces crushed and deformed to the point the joint fractures on the spot.

Beyond these instant failures, there are others that are more subtle or take longer to manifest. Overtightening can cause bolt relaxation, an initial response to preload that’s related to the primary stages of creep, to accelerate. In short, the bolt shank is stretched and loses elastic recovery, permanently lengthening, which causes a loss of preload tension in the bolted joint. Overtightening a bolted joint may also cause thread galling.

All of this is a long way of saying that torque is a vital component of preload, something that absolutely needs to be considered and understood to get the preload on bolted joints right, which is why we’re taking the time to explore it.

Torque and Preload in Bolted Joints

When you tighten a nut or a bolt, you’re applying a twisting force called torque to it. Most people with an engineering background are perfectly comfortable with torque, but what's less widely appreciated is how little of it actually reaches the bolt as useful work.

Of the torque you apply, roughly 50% is swallowed by friction where the nut or bolt head bears against the surface it turns against. Another 35 to 40% is lost to friction in the threads. That leaves only about 10 to 15% doing the real job: stretching the bolt to create preload.

So if you apply 100 Nm of torque with your spanner, only about 10 to 15 Nm worth of that is genuinely creating preload. The rest is just heat from friction.

The relationship between torque and preload is explained with the following bolt preload formula:

 

T = K × F × d

 

If this is the first time you’ve seen this equation, here’s what the different parts of it mean:

• T is the torque you apply 

• F is the preload you get (the tension in the bolt, which is the thing you actually want)

• d is the bolt diameter (a bigger bolt means more torque needed for the same preload)

• K is the nut factor, which is the tricky one

K is a single number that bundles up all the friction effects. A dry, unlubricated steel bolt might have a K of around 0.20, but if you add a bit of oil it drops to about 0.15. Use a specialist lubricant like molybdenum disulphide and it can go as low as 0.10.

Here's why that matters so much. If you apply the same torque to the same bolt but change the lubrication, you get a completely different preload. A bolt torqued to 50 Nm with a K of 0.20 will have roughly half the preload of the same bolt torqued to 50 Nm with a K of 0.10.

This is why torque alone is an imperfect way to control bolt preloading. You're not measuring the thing you actually care about (how much the bolt is stretched). You're measuring the input (how hard you twisted) and hoping the friction conditions are what you assumed. The typical accuracy of torque-only tightening is about 25 to 35%, which means your actual preload could be anywhere in quite a wide band either side of your target.

It works well enough for most everyday applications, but it's why critical joints often use more precise methods and why using the right lubricant (or no lubricant, if the torque spec was set for dry conditions) is so important. Swapping the lubrication without adjusting the torque value is one of the most common ways bolted joints get dangerously over- or undertightened.

The Factors That Affect Preload and Clamping Force

Lubrication is only part of the story. Several other factors influence how much preload and clamping force you end up with and it's worth knowing what they are.

Friction

Friction does more than resist the threads. It actually shows up in two places: in the threads themselves and at the bearing face where the nut or bolt head rotates against the joint surface. Both draw off torque that would otherwise become preload.

The condition of those surfaces affects the preload and clamping force. A corroded thread or bolt generates additional friction. So does a fastener that's been reused, because tightening deforms the contact surfaces at a microscopic level. Even coatings and platings change the behaviour of the joint.

A washer between the fastener and the joint surface adds another friction interface. A corroded washer will generate more friction than a new one with smooth, even surfaces. Similarly, a plastic washer will have a different friction coefficient than a metal one and will interact with the joint surface differently.

If a prevailing torque locking nut is used, a portion of the torque being applied to the bolted joint in order to fasten it will be used to deform the thread. This is normal, it’s how lock nuts work, but worth remembering that the torque needed to fasten one of these fasteners will be affected and a portion of it won’t affect the preload or clamping force.

Material Properties and Bolt Grade

Friction aside, the materials themselves shape how much bolt preloading and bolt clamping force a joint can take.

The bolt's material sets the ceiling for how much preload you can safely generate. Every bolt has a proof load, which is the maximum tension it can sustain without permanently deforming. Go beyond it and the bolt permanently stretches, losing its spring-like elasticity. Once this happens, it can't maintain preload properly anymore.

The standard target for most general-purpose joints is about 75% of the bolt's proof load. In more controlled environments where friction is well characterised and tightening methods are more precise, engineers might push to 90%. Going higher than that leaves almost no safety margin before the bolt yields.

Steel bolts are graded by strength. The numbers you see stamped on bolt heads (8.8, 10.9, 12.9) tell you how much load the bolt can handle. The first digit is the bolt's nominal tensile strength in hundreds of megapascals (MPa), so an 8.8 is rated at around 800 MPa, a 10.9 at 1000 MPa and a 12.9 at 1200 MPa. The second is the ratio between yield and tensile strength, so the .8 in an 8.8 means the bolt reaches its yield point at about 80% of that tensile figure, or roughly 640 MPa. The higher the number, the stronger the bolt. A 12.9 bolt of the same size as an 8.8 can take significantly more preload before it runs into trouble, which means it can generate more clamping force.

But stronger isn't automatically better. A higher-grade bolt has less room for error. If you overshoot the torque on an 8.8, you've got a wider margin before something goes wrong. Do the same on a 12.9 and you're working much closer to the bolt's limits. The extra strength only pays off if your tightening process is precise enough to take advantage of it without overshooting.

The clamped materials matter just as much. If the parts being squeezed together are softer than the bolt (aluminium housings, plastic components, thin sheet metal), the clamping force can damage or permanently deform the joint faces before the bolt reaches its target preload. 

In those cases, the joint's compressive strength becomes the limiting factor, not the bolt's tensile strength. Washers help distribute that clamping force over a wider area and protect softer surfaces, though remember that if you use them, they will affect the friction generated when fastening.

Joint Stiffness

Joint stiffness refers to how the bolt and the clamped parts share any external load that gets applied after assembly.

Think of the bolt as a relatively flexible spring and the clamped parts as a much stiffer spring. In a typical steel joint, the clamped material might be between 3 and 5 times stiffer than the bolt. When an external tensile load hits the joint (something trying to pull it apart), that load gets shared between the two springs.

Because the clamped parts are stiffer, they absorb most of the external load by releasing a small amount of their compression. The bolt only picks up a small additional tension on top of its preload. This is enormously beneficial for fatigue life because the bolt sees only a small fluctuation in load with each cycle rather than the full magnitude of the external force.

This is also why preload is so critical. If there's insufficient clamping force and the joint separates then the bolt suddenly sees the entire external load directly. The stress fluctuation jumps dramatically and fatigue failure follows quickly.

Joint stiffness can be influenced by design. A longer bolt with more grip length is more flexible, which sounds bad but actually means it's more tolerant of external loads and relaxation because it takes more stretch to change its preload by a given amount. Using a proper washer or increasing the footprint of the bolt head helps the clamping force spread through a wider cone of material, effectively making the joint stiffer. Both effects improve the ratio in the joint's favour.

Embedment and Relaxation

A freshly tightened bolt will lose some of its preload in the minutes and hours after assembly, even with no external load. 

At a microscopic level, the surfaces in contact (thread flanks, bearing faces, the interfaces between clamped parts) are never perfectly smooth. When the bolt is first tightened, the high points on these surfaces compress and flatten under the clamping force. As they settle, the overall grip length shortens by a tiny amount. Because the bolt was stretched to a specific length, any shortening of the stack means the bolt relaxes slightly and preload drops. A typical loss is around 5 to 10% of the initial preload.

Gaskets make this worse. Gasket materials are designed to deform to create a seal, so they compress and creep significantly more than metal-to-metal interfaces. Gasketed joints almost always need re-torquing after initial settling.

Temperature Effects

Every material expands when heated and contracts when cooled, but different materials do it at different rates. This comes into play whenever the bolt and the clamped parts are made of different materials.

Take a common scenario: a steel bolt clamping an aluminium housing. If the assembly heats up during operation, the aluminium expands more than the steel bolt. Because the bolt is constrained by the nut, the extra expansion of the aluminium effectively stretches the bolt further, so preload and clamping force increase. If the temperature rise is large enough, the bolt can yield, so when the assembly cools back down the bolt will be loose because it's now permanently stretched.

The reverse happens on cooling. The aluminium contracts more, the effective grip length shrinks, so preload drops. In extreme cases or after thermal cycling, the joint can end up with almost no clamping force at all as a result of this.

Methods of Controlling Preload and Clamping Force

Now that we know what preload and clamping force are, what happens if you get them wrong and the different factors that can affect them, it’s time to look at how you control them. There are several different ways to manage them and ensure you get them right, so take a moment and consider which of these might be best for your application and workflow:

 

Note that accuracy values are indicative ranges based on widely published engineering references. Actual accuracy depends on application-specific conditions, including friction control, operator skill, equipment calibration and joint design.

Common Mistakes and How to Avoid Them

By now the theory is in place. Engineers know what preload and clamping force are, what affects them and how to control them. Yet bolted joints still fail all the time and more often than not it comes down to a handful of simple, avoidable mistakes rather than anything complicated. Here are the seven that catch people out most often, and how to steer clear of them.

1: Reusing a torque spec with a different lubricant or coating

This is probably the single most common way joints end up dangerously over- or under-loaded and it happens because it feels like nothing has changed. The bolt is the same, the wrench is the same and the torque value is the same. But if the original spec was set for a dry bolt and you apply a thread lubricant, you've dropped the nut factor significantly. That means far more of your torque is converting into preload than the spec intended, so you could easily push the bolt past its proof load and into yield or strip the threads out of a tapped hole. 

Going the other way is just as risky. If the spec assumes lubricated threads and you install dry, you'll fall well short of the target preload and end up with a joint that doesn't have enough clamping force to stay tight under service loads. 

The fix is straightforward: always check whether the torque value you're using was intended for the specific friction conditions you've got. If the lubricant, coating or plating has changed, the torque value must change with it.

2: Single-pass tightening on multi-bolt joints

When you tighten the first bolt in a flanged joint or a cover plate, it pulls the joint faces together in its immediate vicinity. But the flange isn't perfectly rigid, so it flexes slightly. When you move on and tighten the second bolt, the flange distorts again and some of the clamping force from the first bolt is partially relieved. 

By the time you've worked your way around every bolt, the ones you tightened first are sitting well below their target preload, resulting in uneven clamping force across the joint. This causes localised overloading, uneven gasket compression and potential leaks. 

The solution is to tighten in multiple passes using a star or cross pattern. Do three or four passes, incrementally increasing the torque each time to something like 30%, 60%, 100%, then a final check pass. Each pass brings the overall distribution closer to uniform. It takes longer, but on any joint with more than four bolts it's the only way to get consistent clamping force across the assembly.

3: Confusing proof load with yield strength

These two terms sound similar and they're often used loosely, but they're not the same thing. Proof load is the maximum tensile force a bolt can sustain without taking any permanent deformation. Yield strength is the point at which the material begins to deform plastically. 

Proof stress sits below yield stress (for an 8.8 bolt, proof stress is 580 MPa against a yield of 640 MPa). If you set your target preload based on a percentage of yield strength rather than proof load, you're aiming higher than you should be and you've eaten into the safety margin that accounts for friction scatter, relaxation and the accuracy limits of your tightening method. 

Always reference proof load when calculating target preload. It's the number that already has the right level of caution built in.

4: Reusing torque-to-yield bolts

Some bolts are designed to be tightened past their elastic limit into plastic deformation. This squeezes out the uncertainty of friction because once the bolt yields, further rotation produces a very consistent and predictable clamp. These are called torque-to-yield bolts, automotive cylinder head bolts and connecting rod bolts are common examples of them in action. 

The catch is that the bolt has now permanently stretched. Its elastic behaviour has changed, it's lost some of its ability to maintain preload and its fatigue resistance has been reduced. These are single-use fasteners by design, so reusing them means you're starting from an unknown baseline.  The bolt might yield almost immediately, it might not generate adequate clamping force, but in either case its remaining fatigue life is unpredictable.

5: Relying on operator feel instead of calibrated tooling

An experienced fitter can get surprisingly close to a target torque by feel on smaller fasteners. The problem is that "surprisingly close" still means a scatter band far wider than any torque wrench would produce and it's completely unverifiable. There's no record, no repeatability and no way to audit the joint afterwards. 

On critical joints the risk is obvious, but even on routine assembly work the cumulative effect of loose preload control adds up. You end up with inconsistent clamping force across a machine or structure, which means some joints are carrying more load than they should be, while others aren't carrying enough. 

Calibrated tooling doesn't have to mean expensive. A basic click-type torque wrench, properly calibrated and stored correctly, removes the guesswork. For higher-volume or higher-criticality work, electronic torque wrenches that log every tightening event add traceability as well as accuracy. 

6: Ignoring relaxation and not re-torquing after initial settling

As we mentioned earlier, a freshly tightened joint will lose 5 to 10% of its preload as surfaces settle, coatings compress and rough surface asperities flatten. In gasketed joints, the loss can be significantly higher. 

If the initial torque is applied and the joint is immediately put into service without a re-check, the actual clamping force in operation is lower than the assembly torque implied. For many joints, this settling loss is within the design margin and doesn't cause problems. But for gasketed connections, joints in vibrating equipment or anything where the clamping force margin is tight, a re-torque after a short settling period is essential. 

Some critical applications specify a re-torque interval in the maintenance schedule for exactly this reason. The time and cost of a re-torque pass is negligible compared to a leak, a loosened joint or a fatigue failure.

7: Mixing bolt grades in the same joint

If you've got six bolts in a flange and five are class 10.9 but one has been swapped out for an 8.8, you've got a problem. That 8.8 bolt is weaker, it can't handle as much preload as the others. Torque the whole joint to a 10.9 spec and that one bolt is being asked to do more than it's built for, meaning it might yield, lose preload over time or might fail outright.

Either way, the clamping force across the joint is now uneven. The remaining five bolts are picking up the slack, which means they're carrying more load than they were designed for too. 

It works the other way as well. An over-spec bolt won't necessarily fail, but it introduces inconsistency, which is what causes unpredictable joint behaviour.

Every bolt in a joint should be the same grade, the same material and the same coating. This is where sourcing from a quality supplier matters. Consistent, correctly marked, certified fasteners mean you know exactly what's in every position and you can trust that they'll all behave the same way when you tighten them.

Key Takeaways

Preload and clamping force in bolted joints are linked, meaning you can’t have one without the other. Preload is what holds a bolted joint together, not torque. That said, there are a lot of different factors that can influence the amount of torque needed to generate the necessary preload in a bolted joint, like the use of different washers and the materials present in the fasteners and joint itself. Friction is the single biggest factor affecting preload, so understanding this is vital to ensuring you don’t under- or over-tighten the joint.

If you’re still at a loss as to what you need to correctly bolt your joints, feel free to contact Accu’s team of qualified and experienced engineers. They can advise you on what you need to do to ensure the right preload is present in your bolted joints, as well as steer you to the right grades of nuts and bolts you need for your assembly.

Further Reading:

• Nuts and Bolts Suppliers UK: How to Source Quality Fixings for Any Application.

• Stainless Steel Fasteners: Recommended Tightening Torques.

• Fasteners & Vibration

FAQs:

Q: What is the difference between preload and clamping force?

A: They're the same force seen from two perspectives. 

• Preload is the axial tension in the bolt, pulling the bolt head and nut towards each other. 

• Clamping force is the equal and opposite compression acting on the parts sandwiched between them. 

If there's 10 kN of preload in the bolt, there's 10 kN of clamping force on the joint. Engineers tend to talk about preload when they're concerned with the bolt (will it yield or fatigue?) and clamping force when they're concerned with the joint (will it slip, seal or separate?).

Q: How much preload should a bolt have?

A: The standard target for most general-purpose joints is around 75% of the bolt's proof load. In more controlled environments where friction is well characterised and tightening methods are precise, engineers may push to 90%. Going higher leaves almost no margin for friction scatter, relaxation or tightening accuracy before the bolt yields. 

The target should never exceed proof load and the clamped materials need to be able to withstand the resulting clamping force without deforming.

Q: Why does a bolt loosen even when tightened to the correct torque?

A: The most common cause is transverse vibration. Even small side-to-side movements can overcome thread friction and allow the nut to rotate incrementally in the loosening direction, a mechanism known as the Junker effect. Other causes include embedment relaxation (surface settling that reduces preload by 5 to 10% shortly after assembly), insufficient initial preload that leaves the joint vulnerable to slip, thermal cycling that alters the clamping force and gasket creep in sealed joints. 

In many cases the issue isn't that the torque was wrong but that the resulting preload wasn't high enough to keep the joint faces locked together under the service conditions.

Q: Does lubricating a bolt increase preload?

A: Yes, for the same applied torque. Lubrication reduces the nut factor (K), which means less torque is lost to friction and more converts into useful preload. A dry steel bolt with a K of 0.20 will produce roughly half the preload of the same bolt lubricated to a K of 0.10, at the same torque setting. 

This is why matching the lubricant to the torque specification is critical. If a torque value was derived for dry conditions and you apply lubricant without adjusting the torque down, you risk overloading the bolt past its proof load.

Q: Can you measure bolt preload directly?

A: Yes, though not all methods are equally practical. The most direct approach is measuring the bolt's physical elongation with a micrometer before and after tightening, since stretch is directly proportional to preload. Ultrasonic systems achieve the same thing by timing a sound pulse down the bolt shank, with accuracy in the range of 1 to 5%. Load-indicating washers (DTIs) provide a go/no-go confirmation that a minimum preload has been reached and load-indicating bolts with visual feedback (colour change or flush pins) offer at-a-glance verification. 

Torque alone does not measure preload directly. It measures the input force and relies on assumptions about friction to estimate the output.

Q: What does torque-to-yield mean?

A: Torque-to-yield (TTY) is a tightening method where the bolt is deliberately taken past its elastic limit into plastic deformation. The bolt is first snugged to a specified torque, then rotated by a further angle that pushes it beyond yield. 

Once the material yields, further rotation produces a very consistent and predictable preload regardless of friction, which is why the method is used in high-performance applications like automotive cylinder heads and connecting rod assemblies. The trade-off is that the bolt has permanently stretched and cannot be reused. TTY bolts are single-use fasteners by design.