Half of the samples were annealed with the initial torsion kept intact and half with the torsion released before annealing. Releasing the torsion meant letting the line unwind to an equilibrium state after the initial forming of the coils by twisting. This caused the sample to shorten as the coils grew in diameter and “loosened”. The samples are annealed under tension at 250g for 30 minutes then allowed to cool. Once cool, the fibers were then treated again under heating and cooling cycles to remove any hysteresis with approximately 250g weight. Once enough viable samples were created for testing, the coils were tested for tension and basic durability in a device.

The chart to the left shows the temperature versus length relationship on the coiled nylon filament as it is being annealed. The orange shows the filament that has the excess torsion released before starting the annealing process and the blue shows the filament with the excess torsion kept during the annealing.

The photos above shows the angles of the two filaments. The filament with a 65˚ angle was annealed with the torsion. The filament with the 70˚ angle was annealed without the excess torsion.

The chart showing the rate of coiling compares the number of rotations needed to coil a 50cm long filament. The dip seen in the graph between the blue and grey lines is the onset of coiling- when the torsion becomes high enough that the filament starts to self-coil.

The orange line represents the calculated progression for coiling. It stops at the point the computations reach a negative number since the formula uses a square root.


Where 500 is the length in micrometers, and pi is 3.145... as computed by a computer.