2.3.1. Measurement System Description

Bilaterally ankle torque measurements were performed for all participants using a reliable [41], reproducible [42], portable custom-made electronic dynamometer [44]. A simple diagram with the measurement system components is represented in Figure 1.

The dynamometers pedal construction permitted the measurement of ankle torque at different joint angles by setting the desired pedal angle using an electronic inclinometer. Torque was converted into force, permitting that the applied force on the load cell was to be further converted into voltage. By the connected load cell amplifier, the load cell imbalance was converted into voltage, further evaluated with an oscilloscope (PicoScope Model 2204A) [45] connected to the personal computer (PC). The PicoScope®6 software used (manufactured by Pico Technology, PC Oscilloscope software version: 6.14.54.6108Copyright © 1995–2021, Pico Technology Ltd., St Neots, UK) [46] permitted the data acquisition and the recording of the whole measurement period. The apparatus construction, calibration and the full disclosure of the measurement system and protocol intervention have been described elsewhere [41]. A representation of the measurement system components and the participant's general position on the chair is captured in Figure 2. Figure 2 demonstrates the minimum required amount of space (~3 m2) needed for the whole measurement system components and the participants' chair to be positioned in order to proceed to measurements.

**Figure 1.** Diagram of graphic representation of the measurement system components comprising the portable custom-made electronic dynamometer, dynamometer load cell, a load cell amplifier connected to the dynamometer's load cell through a four-wire cable, the USB connected oscilloscope (PicoScope 2204A) and the personal computer (PC).

**Figure 2.** The system components used for ankle torque measurement and participant position on the chair; the measurement system components include: portable custom-made electronic dynamometer with an incorporated load cell, electronic inclinometer, a load cell amplifier, oscilloscope, connection wires and a PC.

The PicoScope®6 software graphic user interface allowed for the particular parameter configuration seen in Figure 3.

**Figure 3.** Flow chart representing the steps needed for the selection of the oscilloscope software graphic user interface configuration parameters: Channel A on, direct current (DC) coupling, input = 2 V/div, time base = 100 ms/div (32 s length record) and 10 bits resolution enhancement.

During active ankle plantar flexion, a positive voltage value variation was obtained, as seen in Figure 4—trace A, while during dorsiflexion, a negative voltage value variation was obtained, as seen in Figure 4—trace B. The voltage values were transformed into torque in Nm. The actual voltage value, represented in V, corresponds to the voltage signal seen on the PC screen. Two sources of displacement of the voltage off-set are present in the set-up: the pedal remanent off-set (own pedal weight and permanent mechanical tension), as seen in Figure 4—trace E, and the participant-generated off-set, as seen in Figure 4—traces C and D (comprised of the lower limb weight in the absence of any ankle motion without tension generated by the fixation belt, and the lower limb weight in the absence of any ankle motion with the tension generated by the fixation belt, respectively).

After each measurement, the data recordings were saved as a folder from the PicoScope interface, containing 32 text files of voltage values. For the validation of each measurement, a MATLAB [47] application was developed. The application allowed for the inspection of the saved text files by loading the multi-file contents of each measurement and concatenating them in a single time graph. By inspection of the resulted time graphs the operator appreciated the quality of each measurement and concluded for validation. Time graphs not passing the validation procedure were followed by another measurement trial. Measurements considered valid received a numerical code for each participant. The saved valid measurements were later sent for data processing. For accurate time graphs, low-pass filtering and scaling with the pedal constant were applied [41]. Voltage offset and peak torque during MVIC were processor-estimated and summarised into an Excel spreadsheet [48]. The summarised Excel data of voltage were converted into torque data [41]. Both time graphs of the voltage data, as well as time graphs of torque data, were visualised as graphic representations in time of the voltage variations in V, as seen in Figure 5a, or torque variations in Nm, as seen in Figure 5b–d. Automatically computed peak torque values (highlighted with a red circle, as seen in Figure 5c,d) and offset means (indicated by the red line in the time graphs, as seen in Figure 5c,d) were included in the torque time graphs.

**Figure 4.** Time graph representing five different situations. Trace A represents a positive transition of voltage from the off-set level during two MVICs while the participant actively plantarflexes the foot; Trace B represents a negative transition of voltage from the off-set level during two MVICs while the participant actively dorsiflexes the foot; Trace C represents the off-set voltage level while the participant's foot is relaxed on the pedal with fixation strap not tightened; Trace D represents the off-set voltage level while the participant's foot is relaxed on the pedal with the fixation strap being tightened on the thigh just above the knee level; Trace E represents the pedal off-set in the absence of the participant's foot on the dynamometer pedal.

**Figure 5.** Three time graphs representing a succession of three MVIC of 5 s each followed by 5 s of relaxation between contractions: (**a**) time graph of voltage during ankle plantar flexion with the period being represented in seconds (s) on the Ox axis and voltage represented in V on the Oy axis; (**b**) an example of the same time graph for ankle dorsiflexion with the period being represented in seconds (s) on the Ox axis and torque represented in Nm on the Oy axis; (**c**) torque represented in Nm during ankle plantar flexion, where the peak torque is highlighted with a red circle and the mean off-set level is marked with a red line; (**d**) torque represented in Nm during ankle dorsiflexion, where the peak torque is highlighted with a red circle and the mean off-set level is marked with a red line.
