The collar is unthreaded, so the bolt cuts its own path through it and the polymer grips the thread.

That grip resists rotation in both directions, which is why running the nut down takes noticeable effort from the moment the bolt reaches the insert.

The friction is prevailing torque - it exists regardless of how tight the joint is.

So the nut still resists loosening even if the joint's clamping load is lost.

That is the property a plain nut with a spring washer does not have.

It is also why the bolt must project fully through the insert for the lock to engage.

Considerably less than an all-metal nut - the manufacturer states a maximum, and it is a hard limit.

The polymer softens as temperature rises and loses its grip on the thread.

Past its limit it degrades permanently, so the nut is finished even after it cools.

A nut that has been through a high temperature may look perfectly intact and lock nothing.

Engine bays, exhaust systems, ovens and any process heat rule them out.

Distorted thread and flexible top nuts are the all-metal answers where temperature governs.

A small number stated by the manufacturer - and the field test is whether it still resists by hand.

Each installation deforms and shaves the insert, so the grip reduces every time.

If the nut runs freely down the thread by hand before it reaches the joint, the insert has stopped working.

That check takes a second and should be habitual on any disassembly.

A nut that has been heated is scrap regardless of how many times it has been used.

They are inexpensive enough that replacement at every disassembly is the normal practice.

Because the insert is above the threaded portion - a bolt that stops short of it engages nothing.

The nut will feel tight, because the joint is clamped by the threaded part of the nut.

But no locking has taken place at all, and the assembly is a plain nut with extra height.

This is a common and completely invisible assembly error.

The check is that thread is visible above the nut, or at minimum flush with the insert.

It is one of the reasons bolt length is specified rather than left to whatever is in the box.

Solvents, fuels, some cleaning chemistries and prolonged ultraviolet exposure - all things that do not touch steel.

The nut can be mechanically sound and have lost its lock to its environment.

Nylon also absorbs moisture, which changes its dimensions and its grip.

Washdown environments with aggressive detergents are a common and under-appreciated case.

Outdoor exposure over years embrittles unstabilised polymer.

Where the chemistry is uncertain, an all-metal locking nut removes the variable entirely.

Yes - the prevailing torque of the insert has to be added to the torque that produces the clamping load.

Some of the applied torque is consumed overcoming the insert before any clamping begins.

Using a plain-nut torque figure therefore under-tightens the joint.

Manufacturers publish prevailing torque values, and the assembly torque is the sum of the two.

Prevailing torque also falls with each reuse, so the sum changes as the nut ages.

Where preload is critical, angle-controlled or tension-indicating methods avoid the issue.

High temperature, aggressive chemistry, frequent disassembly and safety-critical joints that need a positive lock.

Temperature and chemistry both attack the insert directly.

Frequent disassembly consumes the insert quickly and produces nuts that look fine and lock nothing.

Where a lock must be positive rather than frictional, a slotted or castle nut with a cotter pin is the correct choice.

Aerospace and lifting applications frequently specify positive locking for that reason.

For ordinary vibrating machinery within temperature, they remain the sensible default.