Torque vs Preload: The Fundamental Relationship

When you turn a bolt with a torque wrench, the torque you apply does two things. A small portion stretches the bolt and creates the clamping force (preload) that actually holds the joint together. The much larger portion is consumed overcoming friction in the threads and under the bolt head or nut face. The relationship is conventionally written as T = K x F x d, where T is the applied torque, K is the torque coefficient, F is the desired preload, and d is the bolt nominal diameter.

The torque coefficient K is the critical number. For an unlubricated, zinc-plated steel bolt, K typically lands between 0.18 and 0.22. For a lubricated bolt, K can drop to 0.10 or lower. This means the same applied torque produces nearly twice the preload on a lubricated bolt compared to a dry one. That is why a torque chart that assumes a dry, plated surface will badly under-torque a lubricated fastener, or over-torque a dirty, rusty one. Per GB/T 16823.2-1997 (general rules for the tightening of threaded fasteners), torque control typically targets about 70% of the proof stress Sp; for an 8.8 M12 bolt the proof stress is roughly 580 MPa and the tensile stress area, taken per GB/T 196, is about 84.3 mm², giving a target preload of around 34 kN.

Recommended Torque Reference (8.8/10.9/12.9)

The table below gives typical target torques for common bolt sizes at a 0.14 torque coefficient (clean, zinc-plated, dry), targeting approximately 70-80% of yield. Always verify against your actual friction coefficient and joint requirements.

These values are starting points. If your bolts are waxed, oiled, or have a specific friction-control coating, the torque must be adjusted using your measured K factor. For safety-critical joints, do not rely on a chart; run a torque-clamp load test per GB/T 16823.3.

Friction Coefficient: Why Torque Charts Are Not Exact

Friction coefficient has two components: thread friction and bearing-surface friction. Thread friction occurs between the male and female threads as the bolt advances. Bearing-surface friction occurs under the rotating bolt head or nut flange. Each component contributes roughly half of the total friction loss, and either one can vary by 50% or more depending on surface finish, lubrication, plating, and contamination.

This is why two bolts from the same box, torqued to the same value, can end up with very different clamp loads. If one has residual drawing oil on the threads and the other is dry, the preload scatter can easily be plus or minus 30%. In safety-critical or gasketed joints, that scatter means some bolts are under-clamped and leak, while others are over-stretched and approach yield. The engineering fix is to control friction: specify a defined lubricant or coating, measure the friction coefficient on a torque-angle tester, and torque to a target derived from that measured value.

Why Bolts Loosen Even at Correct Torque

There are three mechanisms of bolt loosening. The first is rotational self-loosening, where transverse vibration causes the bolt to rotate backwards out of the nut. This is the most common and most damaging mode, and it cannot be prevented by simply increasing torque. The second is slackening or embedding, where microscopic surface roughness peaks crush under clamp load and the bolt shortens slightly, losing preload. This happens in the first few hundred cycles and is usually accommodated by re-torquing after run-in. The third is preload relaxation due to creep or stress relaxation, common in high-temperature joints or plastic-bolted joints.

If a bolt loosens despite correct torque, the root cause is almost always one of these three rather than "not tight enough." Increasing the torque beyond the yield limit does not stop rotational loosening; it only risks yielding the bolt. The correct fix is an anti-loosening device: prevailing-torque nuts, pre-applied thread adhesive, spring or wedge washers, or a direct tension indicating system.

Field Remedies for Stripping and Galling

Thread stripping happens when the internal thread strips before the bolt breaks in tension. It is usually caused by too short an engagement in a soft material (aluminum, magnesium, plastic), by a nut that is too weak for the bolt class, or by cross-threading during installation. The fix is to increase engagement length, use a thread insert in soft material, or specify a stronger nut class.

Galling is a seizure phenomenon that happens most often with stainless steel fasteners under high friction. As the threads slide, local pressure welding occurs and the nut seizes on the bolt, often tearing the threads. Prevent galling by using lubricated stainless fasteners, choosing the correct thread fit, avoiding over-torque, and using anti-seize compound on stainless assemblies. For high-volume automatic assembly, specify a defined friction-control coating rather than relying on dry stainless.

Practical torque checklist for the shop floor. Before tightening, verify that the bolt and nut are the correct property class and that the threads are clean and undamaged. Confirm whether the fastener is lubricated or dry, because the torque chart assumes a defined friction condition. Use a calibrated torque wrench and recalibrate it periodically. For critical joints, use a torque-angle wrench rather than a simple click wrench, because torque-angle controls the bolt stretch more directly. After the first assembly run, check a sample of bolts with a ultrasonic or load-cell measurement to confirm that the target clamp load is actually being achieved. This single check catches the vast majority of systematic torque-process errors.

Understanding the torque-preload relationship in practice. The reason torque control has inherent scatter is that friction coefficient varies with every batch. A bolt that arrives dry from the plater has a different K factor than one that has been sitting in oil for a week. The engineering response is to stop trying to control preload through torque alone and move to torque-angle methods. Torque-angle tightening first brings the bolt to a snug torque, then turns it a specified additional angle. Because the angle directly correlates with bolt stretch, it is far less sensitive to friction variation. On modern automatic assembly lines, torque-angle with transducerized spindles is the standard for critical joints, and it reduces preload scatter from plus or minus 30% to plus or minus 10%. This is why the investment in a better tightening strategy almost always pays for itself in reduced warranty and rework costs.

Final thought on torque process. The most robust tightening strategies in modern manufacturing move away from pure torque control toward torque-angle, yield-controlled tightening, or direct tension measurement. Yield-controlled tightening drives the bolt to just below its yield point, which maximizes preload consistency and uses the bolt's capacity most efficiently. For high-volume automotive and battery assembly, this is the state of the art. For general industrial equipment, a well-calibrated torque wrench with a defined friction coefficient is adequate. The key is to choose the tightening method that matches the joint's criticality, and to verify it with measured clamp load rather than assuming the chart value is correct.

Practical torque checklist for the shop floor. Before tightening, verify that the bolt and nut are the correct property class and that the threads are clean and undamaged. Confirm whether the fastener is lubricated or dry, because the torque chart assumes a defined friction condition. Use a calibrated torque wrench and recalibrate it periodically. For critical joints, use a torque-angle wrench rather than a simple click wrench, because torque-angle controls the bolt stretch more directly. After the first assembly run, check a sample of bolts with a ultrasonic or load-cell measurement to confirm that the target clamp load is actually being achieved. This single check catches the vast majority of systematic torque-process errors.

Understanding the torque-preload relationship in practice. The reason torque control has inherent scatter is that friction coefficient varies with every batch. A bolt that arrives dry from the plater has a different K factor than one that has been sitting in oil for a week. The engineering response is to stop trying to control preload through torque alone and move to torque-angle methods. Torque-angle tightening first brings the bolt to a snug torque, then turns it a specified additional angle. Because the angle directly correlates with bolt stretch, it is far less sensitive to friction variation. On modern automatic assembly lines, torque-angle with transducerized spindles is the standard for critical joints, and it reduces preload scatter from plus or minus 30% to plus or minus 10%. This is why the investment in a better tightening strategy almost always pays for itself in reduced warranty and rework costs.

Final thought on torque process. The most robust tightening strategies in modern manufacturing move away from pure torque control toward torque-angle, yield-controlled tightening, or direct tension measurement. Yield-controlled tightening drives the bolt to just below its yield point, which maximizes preload consistency and uses the bolt's capacity most efficiently. For high-volume automotive and battery assembly, this is the state of the art. For general industrial equipment, a well-calibrated torque wrench with a defined friction coefficient is adequate. The key is to choose the tightening method that matches the joint's criticality, and to verify it with measured clamp load rather than assuming the chart value is correct.