Affordable Access

Time-based calibrations of pressure sensors improve the estimation of force signals containing impulsive events

Publication Date
  • Physics


Piezoresistive pressure sensors are widely used in biomechanics applications involving both static and dynamic loading conditions. The overall accuracy of these sensors has been reported previously in the literature, and multiple linear or polynomial custom calibrations have been proposed to enhance sensors’ performance mainly in low dynamics conditions. The aim of this technical note was to propose a ‘point-to-point’ time-based method to improve Tekscan F-Scan sensor calibration procedures in reconstructing a force signal with variable dynamic content and duration, using an application-specific loading pattern, characterised by an initial impact followed by a slow dynamic phase. The performance of the proposed calibration procedure was compared with four methods divided into time-based calibrations (‘point-to-point’, ‘log time–based’ and Tekscan ‘step’) and linear calibrations (‘drop-ball’ and Tekscan ‘single-point’). The ‘point-to-point’ calibration was the only method providing accurate force estimation over the entire duration, showing an inaccuracy of about 10% both in impact and slow dynamic phases. Tekscan default calibrations (‘step’ and ‘single-point’) underestimated the criterion force by ∼60% over the impact phase but performed better in the slow dynamic phase (∼20% of inaccuracy). ‘Log time–based’ and ‘drop-ball’ performed well during the impact phase (∼11%) but overestimated the slow dynamic phase by ∼170%. For this reason, we recommend ‘point-to-point’ calibration for estimation of forces which are characterised by an initial impulsive event and a subsequent slow-changing load. These findings highlight the importance of selecting the most appropriate calibration with respect to signal dynamics, in terms of loading range, loading pattern and impact duration.

There are no comments yet on this publication. Be the first to share your thoughts.


Seen <100 times