Detecting touch-down, calculating accelerations and strain

Touch-down (TD) was the point in time when the frame made contact with the polystyrene. It was determined in each trial as the point before the earliest instant at which the second time derivative of the COM position (acceleration) had raised above noise level. As no delay between the acceleration signals of frame and COM was detectable in our experiments around TD, either signal could be used in principle to determine TD. However, detecting TD with the COM acceleration was favourable because this method proved more reliable than using frame marker acceleration. Although allowing an extended template size was possible for frame marker tracking (using ‘WINanalyze’, Mikromak Service, Berlin, Germany), and this setup could achieve a spatial noise of just 0.1 pixel for this frame marker, using an array of belly markers (individual noise: 1 pixel) yielded a slightly better signal-to-noise ratio for the array than for the frame marker.

The instants at which maximum accelerations occurred were 7.4 ± 1.0 ms and 7.7 ± 0.9 ms for the frame and COM, respectively. While the COM signal lagged the frame signal by 0.3 ms, on average, a t-test showed that this lag was insignificant at p > 0.05. The maximum COM acceleration values a_COM,max were 165 ± 23 m s⁻² (see right ordinate Fig. 7), which were reached, on average, 2.7 ms earlier than maximum strain. However, strain kinematics are much more variable (see standard deviation of 2.5 ms): in some trials, maximum strain was even reached before maximum COM acceleration, particularly in non-fatigued muscle with low strain maxima (Fig. 4, top). The impact duration measured as the time spent from TD to frame (or bone) acceleration returning to zero was 10.7 ± 0.9 ms. Zero COM acceleration occurred at 11.3 ± 0.7 ms, and the delay to zero frame acceleration was doubled to 0.6 ms as compared to instants of their maxima, which was significant on a level p < 0.05 using a t-test.

To determine belly strain ϵ_CE, an upper and lower range of each ~10% of total muscle length was identified. The vertical placement of both marker subarrays were nearly symmetrically positioned around the midpoint of the belly (location of maximum cross-sectional area A _CE,0,max), as seen in Fig. 1B. The horizontal, white lines across the belly represent the subarray limits and thereby confine what we denote as the ‘contractile element’ (CE) in this study. The representative vertical position of each marker subarray (y _u and y _l with u for ‘upper’ and l ‘lower’) was calculated as the arithmetic mean of the vertical positions of all markers in this subarray. The reference length L _CE,0 that defined zero percent strain is the distance between the vertical subarray positions LCE,0=|yu−yl| measured at TD in each trial (L _CE,0 = 18 ± 0.8 mm). In Fig. 1B, the aponeuroses extend on both lateral sides of GAS, i.e., the field of view. Thus, by solely using markers from the centre of the y _u and y _l regions, care was taken to analyse the kinematics of fibres alone rather than any aponeurosis material.