Braid for vibration isolation

What the braid is doing

A braided connector carries current between two points that move relative to each other. The movement may come from a motor, a compressor, a transformer, a vehicle chassis or a door. A rigid bar would transfer that movement into the terminals, so the braid provides a flexible length that absorbs it.

The braid does not stop the vibration. It decouples the electrical path from the structure, so the terminals are not loaded by the relative movement. The vibration still exists, and the braid still moves. The design question is whether the braid can move for the life of the equipment without breaking strands or loosening a joint.

That makes the braid a fatigue component, not just a conductor. It needs a current rating, a voltage rating and a flex life. Most suppliers will quote the first two. The third has to be designed and tested.

Where the movement comes from

The frequency and amplitude of the movement set the test conditions. A transformer core produces a hum at twice the supply frequency, so 100 Hz on a 50 Hz system and 120 Hz on a 60 Hz system. The amplitude is small, often a fraction of a millimetre, but the cycle count over twenty years is enormous.

An engine or a compressor produces a larger amplitude at a lower frequency, with harmonics. A vehicle adds road input across a wide band and occasional shock loads. A cabinet door or a plug-in module produces a few thousand cycles over its life, but with a large hand-driven movement.

Write down the movement in three directions, the frequency range, the expected cycle count and any shock event. A braid that is routed for a 0.2 mm transformer hum is not the same design as one that must survive a 10 mm engine movement.

What decides flex life

Braid fatigue is driven by the strain in the individual strands. The strand diameter, the number of strands, the braid angle and the free length all change that strain. Finer strands bend more easily, which is why a braid with many fine wires outlasts a rope with fewer coarse wires at the same cross-section.

The free length is the most powerful variable in the design. A longer loop spreads the same movement over more strand length, so the bend angle at each point falls. A short, tight loop concentrates the movement at the terminals and at the first few millimetres of braid.

As an illustrative starting point, allow a free length of at least five times the peak-to-peak movement, then confirm it by test. That is a rule of thumb, not a guarantee. A braid that is also carrying a high current runs hotter, and copper is weaker and more prone to creep at high temperature.

The terminal is the usual failure location because the strands are locked in a crimp or a weld. The movement has to be taken up by the short transition zone just outside the joint. A strain relief sleeve or a longer ferrule spreads the bend over more length, but it must not stiffen the braid so much that the movement moves further along.

Routing rules that matter

Form a single loop in the plane of the main movement. A braid that is asked to bend in two planes at once works the strands against each other and shortens life. If the movement is truly three-dimensional, use a loop with enough slack in all directions rather than a tight S-bend.

Keep the braid in slight compression or neutral, never in tension. A braid under tension has no slack to give, so the first movement loads the terminals directly. The loop should be able to straighten a little without pulling the joint.

Avoid twist. A twisted braid has a different strand geometry and the strands cannot slide as freely. If the terminals are offset, rotate the whole assembly rather than twisting the braid to fit.

Keep the braid clear of sharp edges, hot surfaces and other moving parts. A braid that rubs on a housing will fail from abrasion before it fails from fatigue. A sleeve helps, but it must be sized so that it does not clamp the braid and stop it flexing.

Terminal load and torque

The braid terminal must be clamped so that the strands cannot move relative to the lug. A loose crimp or a bolt that has relaxed lets the strands fret, and fretting breaks strands and raises resistance. Use the terminal that matches the braid construction and follow the crimp or weld procedure.

Bolt torque matters as much as crimp height. A bolted braid lug on a vibrating structure needs a spring washer or a Belleville washer to keep the clamp force as the joint settles. Thread locking compound helps with loosening but does not replace clamp force.

Support the braid near the terminal if the run is long. A clamp a short distance from the lug takes the weight and stops the braid whipping. The clamp must grip the braid without crushing the strands, and it must not create a new hard point where fatigue starts.

Check the clearance between the braid and any live part after the harness is fitted. A braid that touches a busbar under vibration will wear through its sleeve and then through the strands.

Current, heat and derating

A braid in still air runs hotter than the same cross-section on a open busbar because the strands are close together and the surface area is partly shielded. The terminal joints add heat. Size the braid on the actual mounting, not on a catalogue rating for a short free length.

Vibration isolation assemblies often sit inside an enclosure with poor airflow. Measure the braid temperature at the worst-case continuous current and at the highest ambient. Keep the strand temperature below the limit of the sleeve and the terminal insulation.

If the braid is also a flexible connection in a high-current path, check the voltage drop at full load. A braid that is adequate for current but has a long loop may drop more voltage than the system can tolerate, especially at low voltage.

The joint temperature is the number to watch. A thermocouple on the lug and one on the braid body will show whether the heat is coming from the conductor or from the termination. If the lug runs much hotter than the braid, the joint is the problem. If the whole run is hot, the cross-section or the ambient is the problem, and a longer loop will make it slightly worse because it adds resistance.

Failure modes and what they look like

Most braid failures are visible before they are electrical. Broken strands appear at the terminal transition, the braid becomes stiff or kinked, or the sleeve shows a wear mark. The table links the common signs to the cause and the correction.

Inspection and replacement

Inspect the braid at the terminal transition first. Look for broken strands, a loose ferrule, discolouration and fretting marks. Run a hand along the braid to feel for a stiff spot or a kink. Check the sleeve for wear and the mounting clamps for looseness.

Measure the resistance of the assembly and compare it with the build record. A rise of a few per cent is worth investigating. A rise of tens of per cent means the joint or the braid has already lost section.

Replace the braid if strands are broken at the terminal, if the braid is kinked, or if the resistance has moved outside the acceptance band. Do not re-terminate a braid that has already fatigued; the remaining strands have used part of their life.

For a new design, run a flex test at the real amplitude and frequency, with the real terminal and the real mounting. Measure resistance during the test, not only at the end. The resistance curve often shows the first strand breaks long before the braid separates.