EASA / CAR Part-66 · Module 7 · 7.6(b) Fasteners & Torque Loading · Cat A · B1/B2

Torque & Bolt Preload

Objective: a torque wrench does not measure what actually matters. What holds a joint together is the preload stretched into the bolt, and most of the torque you apply is spent overcoming friction before any of it reaches that. See where the torque goes, what preload results, and what happens when the bolt is taken past yield.

Bolted joint, sectioneddrag to rotate

Where the applied torque goes

Thread condition

Applied torque 40 N·m

Bolt size M10

Clamp force (preload)

0kN

Bolt tensile stress

0MPa

Percent of yield

Status

T = K · F · d → F = T / (K · d)

Correctly torqued

The job of a bolt is to be a stiff spring. Tightening stretches it, and that stretch is what clamps the joint. The clamp force is called preload, and it is the only thing keeping the joint faces together and stopping the bolt from seeing the full fluctuating load in service. A properly preloaded bolt in a fatigue-loaded joint barely feels the cycling at all, because the joint takes it; an under-preloaded one feels all of it and fails.

Torque is only a proxy for preload, and a poor one. The relation T = K·F·d contains the nut factor K, which lumps together the thread friction, the friction under the bolt head or nut face, and the thread geometry. On dry threads only around ten to fifteen percent of the torque you apply ends up as useful bolt stretch, rising to perhaps a quarter on well lubricated ones. The rest is burnt as friction. Because K depends on lubrication, plating and surface condition, the same torque on the same bolt can give preloads that differ by a factor of two, which is exactly why maintenance data specifies whether threads are dry or lubricated and why you must never substitute one for the other.

Over-torque is not conservative. Past yield the bolt stretches permanently, loses its spring rate and cannot hold preload. It may feel tight and still be useless, and it will usually fail later under load rather than on the bench. This is why torque wrenches are calibrated, why running down a fastener with an impact driver and finishing by feel is unacceptable, and why critical joints use angle-torque or direct bolt-stretch measurement instead of torque alone.

Try this: Hold the torque fixed and switch from dry to anti-seize. How much does preload change, and what would happen if you used the dry torque figure on lubricated threads?
Try this: Look at the torque breakdown chart. What fraction of your effort actually reaches the bolt, and where does the rest go?
Try this: Wind the torque up until the status turns red. Once a bolt has yielded, why can it no longer hold clamp load even though it still feels tight?