Torque, and the failures you cannot see

This unit is where the curriculum's recurring theme becomes the subject itself.

A fastener torqued to half specification and one torqued correctly look identical. They photograph identically. They pass every inspection that does not involve a wrench. There is no perceptual signature, no leak, no noise, nothing to find.

More elements attach to torque than to any other topic here — K8 torquing tools and procedures, K10 torque and preload, R2 improper torque, R4 torquing critical fasteners, S2 determine and properly torque hardware — and that weight is the ACS telling you where the danger is.

1. Torque is not the point. Preload is.

This is the single most important idea in the unit, and the handbook states K10 without developing it.

A bolt is a spring. Tightening it stretches it, and the stretch produces a clamping force — preload — that holds the joint together. The joint is held by that clamping force, not by the bolt filling the hole. A properly preloaded joint does not move, so the bolt never sees the fluctuating loads that cause fatigue.

Torque is only a way of guessing at preload. It is convenient because you can measure it with a wrench, and it is a poor proxy because most of the torque you apply is spent overcoming friction — under the head and in the threads — rather than stretching the bolt. A small change in friction produces a large change in preload for the same indicated torque.

Three consequences follow, and each is a field practice that looks arbitrary until you see this:

• Lubrication changes everything. Oil, anti-seize or a plated finish reduces friction, so more of the applied torque becomes preload. Applying a dry torque value to a lubricated thread over-stretches the bolt — possibly past yield. If the manual specifies lubricated or dry, it is not a preference.
• Prevailing-torque nuts need their running torque added. A self-locking nut resists turning before it ever touches the joint. That resistance is not producing clamping force, so it must be measured and added to the specified value, or the joint is under-preloaded by exactly that amount.
• A used self-locking nut may not lock. R3, used hardware, is about this: the locking feature is consumed by use, and a nut that spins on freely by hand has none left.

A mechanic who understands preload can reason about a fastener they have never met. One who has memorized a table cannot.

2. Why the extension bar arithmetic lives in mathematics

GEN-H Concepts §3 works the torque wrench extension formula, and this is where it is used. It belongs there because it is algebra and here because it is a fastener.

Set the wrench lower than the target when an in-line extension is fitted. The intuition points the wrong way, which is why it is worth meeting twice.

3. R4 — why some fasteners are different

Torquing techniques on critical, highly-stressed fasteners.

Some joints carry flight loads and some hold a fairing on. The ACS separates them because the consequence of getting them wrong is not comparable, and because critical fasteners often carry additional requirements: a specific sequence, a specific tool, a re-check after a settling period, a witness mark, an inspection sign-off.

Find out which category the fastener is in before you tighten it, not after. The maintenance manual tells you; the fastener does not look different.

4. Safety wire is directional, and beauty is not the test

K5 is the element most often learned as a technique and least often learned as a principle.

The wire must be in tension when the fastener tries to loosen. That is the entire requirement. Everything else — the twist rate, the pigtail, the wire gauge — serves it.

So the question to ask at every installation is: if this bolt starts to back off, does the wire get tighter? If the answer is no, the installation is wrong no matter how neat it looks — and a beautifully executed wire installed backwards is worse than none, because it looks correct and will not be questioned.

Two more points the handbook covers and candidates skip:

• The pigtail is bent back under, so it cannot snag a hand, a glove or a control cable.
• Wire is used once. It work-hardens on installation; reusing it produces a wire that will fail where it was previously twisted.

5. Identification is a safety task, not a stores task

K11 head markings and R3 suspected unapproved parts belong together.

A bolt that fits is not a bolt that applies. This is the same sentence as GEN-B Concepts §1 and GEN-I, arriving in your hand instead of on paper. Aircraft bolts of identical diameter and length differ in material, heat treatment and strength, and the only evidence available at the point of installation is the head marking.

R3 extends this to parts of unknown provenance. A part that is unapproved is not a cheaper version of an approved one; it is a part whose properties nobody has verified. The failure is not that it is bad — it is that it is unknown, and unknown cannot be signed for.

Turnlock fasteners (K4) hold cowlings, inspection plates, doors and other panels that come off often. They are captive: the stud turns in the panel but does not drop out when released. The common three are the Dzus, Camloc and Airloc (FAA-H-8083-30B Ch. 7, Captive Fasteners; Turn Lock Fasteners; Dzus Fasteners; Camloc Fasteners; Airloc Fasteners).

6. What makes a measurement defensible

S14, checking a micrometer for proper calibration, is a small element with large consequences, and it is the one that makes every other measurement in the certificate meaningful.

A measurement is only as good as the instrument, and an instrument's condition is not visible. A micrometer that has been dropped reads consistently and wrongly. A worn anvil, a bent frame, a sticky ratchet all produce confident numbers.

This is why GEN-K R2 and R3 classify instrument use and calibration as risks, and why the equipment and materials specification requires our own instruments to be calibrated with traceable records: a candidate assessed on a measurement taken with an uncalibrated instrument has not been assessed on anything.

The same idea returns in AF-C R4, where maintaining a calibration schedule for cable tension meters is itself an examinable element.

7. The unit divides into identify and make

Identify — welds, cable components, rivets, heat-treated against non-heat-treated alloys, materials by marking, and the selection skills that follow from them. Mostly delivered by imagery, and the reason the equipment and materials specification §B2's specimen library exists: identification cannot be assessed on a single correct example, and serviceable controls must sit alongside defective ones or the candidate is only learning that everything is defective.

Make and measure — safety wire, cable swaging, Vernier and micrometer measurement, concentricity, torque. These need hands and cannot be faked. All of them are in the General bench kit posted to you, instruments included (GEN-E §4).

8. What consumes this unit

License

CC BY-SA 4.0 (Licensing).