Summary

The assessment of hand-transmitted vibration exposure and of potential injuries to the hand-arm system when using hand-held power tools is currently based on the ISO 5349-1 guidelines. The recommended guidelines, however, do not account for the effects of the coupling forces exerted at the hand-handle interface, although many studies have shown the importance of these forces in the transmission of vibration to the hand-arm system. This is partly attributed to the lack of practical methods for mesuring hand-tool interface forces in field applications and in-part to the lack of sufficient data relating injury risks to the applied forces. This study explored a low-cost system for measuring hand-handle interface forces and its feasibility when applied to a hand-held power tool handle. 

The study was conducted in three sequential phases. In the first phase, a measurement system based upon low-cost thin-film and flexible resistive sensors (FlexiForce^®) was developed and the properties of the sensors were explored through systematic laboratory measurements. The sensors could also be trimmed to be adapted to different handles. The properties included the hysteresis, linearity and repeatability of the sensor applied to flat as well as curved surfaces. From repeated measurements, it was observed that the sensors exhibit negligible hysteresis and very good linearity with the applied force. The measurements, however, revealed strong dependence on the loading area, position of the load on the sensor, length of the sensor and flexibility of the loading media (elastomer). Furthermore, considerable differences were observed in the outputs of different sensors, and the sensors showed degradation of the output signal with time and usage. Subsequent measurements performed with 5 different instrumented cylindrical and elliptical handles also showed very good linearity of the sensors but strong dependence on the hand size and the handle size. It was concluded that the sensors could provide reasonably good estimates of the hand forces provided that each sensor was calibrated for the specific handle and hand size.

Optimal locations of the 40 mm wide sensors on the handle surface were determined through the measurement of hand positions on different handle sizes, and from the hand-handle interface force distributions acquired in an earlier study for different sizes of cylindrical handles. It was concluded that two sensors located symmetrically on opposite sides of the handle, in the forearm axis, could provide very good estimates of the palm- and finger-side contact forces. The hand grip and push forces were subsequently derived from the measured palm and finger forces. The validity of the low-cost measurement system was investigated through repeated measurements with 7 subjects and 5 different handles, including three cylindrical handles (32, 38 and 43 mm diameter) and two elliptical handles (32 mm x 38 mm and 38 mm x 44 mm). A LabView program was developed to acquire the FlexiForce^® palm and grip forces, and reference grip and push forces from the instrumented handles, which were displayed to the subject. The experiments were conducted with each subject grasping the stationary instrumented handle with 12 different combinations of hand grip (10, 30 and 50 N) and push (25, 50 and 75 N) forces. The measurements were also repeated under two different levels of broad band vibration in the 4 to 1000 Hz frequency range (frequency-weighted rms acceleration of 1.5 and 3 m/s²). The results showed good linearity and repeatability of the sensors for all the subjects and handles under static as well vibration conditions, but the sensor outputs differed for each handle and subject. This further confirmed the need for calibrating the sensors for each subject and handle.

In the second phase, the feasibility of the sensors for the measurement of the biodynamic response of the hand-arm system was explored. The study was conducted with 6 subjects grasping the 38 mm instrumented handle with nine different combinations of grip and push forces, and two levels of broadband vibration. The handle was also equipped with two FlexiForce^® sensors for the measurements of the palm-handle and finger-handle interface dynamic forces. The data acquired from the instrumented handle were analyzed to determine the palm- and finger-side driving point mechanical impedance responses, which served as the reference values. The driving-point palm- and finger-impedance responses were also obtained from the FlexiForce^® sensors and compared with the reference values to evaluate the feasibility of the measurement system. The comparisons revealed very similar trends, while the impedance magnitude responses of the FlexiForce^® sensors were substantially lower in the entire frequency range, except at very low frequencies. This was attributed to poor frequency response of the FlexiForce^® measurement system. The frequency response characteristics of the sensors were subsequently obtained from the measured responses, which revealed strong dependence on the hand-handle interface forces, handle size and vibration level. The application of a compensation function based upon the measured frequency response characteristics resulted in impedance responses of the FlexiForce^® sensors comparable to the reference values for all the experimental conditions considered in the study. It was concluded that the proposed low-cost measurement system could be applied for measurements of the biodynamic responses and hand forces on real tool handles in the field. The determination of the frequency response functions of the sensors, however, would be quite challenging considering its nonlinear dependence on the hand size, hand forces and handle size. Furthermore, the proposed system eliminates the need for inertial compensation of the measured biodynamic responses, which is known to be a source of error in the reported biodynamic responses of the human hand-arm system exposed to handle vibration. 

Finally, in the third phase, the validity of the measurement system was examined with hand grasping a static as well as a vibrating tool handle under different combinations of hand grip and push forces. The experiments were conducted with a chisel hammer operating in an energy dissipator in the laboratory. Two FlexiForce^® sensors were applied to the primary handle of the tool to measure the palm- and finger-side forces. A methodology was developed to calibrate both sensors. The validity of the measurement system was evaluated with three subjects grasping the static as well as the vibrating tool handle under different combinations of hand grip, push and coupling forces. Measurements revealed very good correlations between the hand forces estimated from the FlexiForce^® sensors and reference values for the static as well as for the vibrating tool. The ratio of the palm force obtained from the FlexiForce^® measurement system to the coupling force ranged from 0.96 to 1.05, when the subjects grasped the vibrating tool handle.  This ratio varied from 0.96 to 1.06 for the stationary tool handle.