Professor Florida Gulf Coast University Fort Myers, Florida, United States
Introduction: Structural testing of lower and upper limb prostheses is essential to provide durable, and safe prostheses for individuals. Standardized mechanical testing for lower limb prostheses involve static and cyclic loading (ISO 10328, 2016). There is not a standardized method of testing for upper limb prostheses as there is for lower limbs. Mechanical testing is often based on the intended function/activity level of the upper limb prosthesis. Typically, the tests performed are derived from other ISO’s and depend on intended wear and tear of components (Mio, Sanchez, & Valverde, 2018). There is additional activity-based testing such as the SHAP, 9-hole peg test, and box and blocks.
Impact tests are performed to the point of failure, which is not optimal for expensive prosthetic components. Beneficial data from an i9-holetest is the ability to see how the prosthetic absorbs shocks/impacts. This project takes an unconventional approach to the mechanical testing of upper limb prosthetics and attachments by creating a non-destructive impact test to view soft vs stiff samples material’s stress responses when struck at varying heights. The development of a non-destructive impact test could improve the design of prosthetic components and material selection to improve the lifetime of the prostheses, attached components, and comfort of the user.
Materials and
Methods: The research project focused on the making and calibration of the impact test shown in Figure 1. Test samples of pine lumber and metal were used to observe how much impact force they absorbed. First, the mechanical test apparatus was created using 2x4 pine lumber, a 16 oz hammer, twine string to help stabilize the sample in a “free state”, and a clamping device attached to the load cell. A stabilizer was 3D printed to attach to the hammer's handle to ensure that the force only struck the material from 1 direction.
The material's behavior was observed using sensors and an ELEGOO MEGA 2560 microcontroller using C++. An accelerometer (MPU6050) was attached to the hammer so that the gyroscope “Y” orientation shows its path as it strikes the test sample. To begin recording data, a trigger was programmed once the gyroscope reached an angular velocity of 0.8 rad/s on the “Y” axis, to log values of the gyroscope, accelerometer, and force readings. The test samples were attached to a clamp that was screwed onto the load cell (WiiFit Sensor). Connected to the load cell was an instrument amplifier (INA-128P), to transmit the load cell data at a fast enough speed to capture the impact. Figure 2 shows the schematic of the electrical components.
Results, Conclusions, and Discussions: Initially, the software needed to be calibrated to get accurate readings of the angular acceleration in rad/sec and force in pounds. To get accurate force readings, an offset and a scale factor needed to be calculated, a 2-point calibration was used with a 0 and 10 lb load applied to the load cell. Once this was resolved, the analog readings were converted to force in pounds.
After calibrating the sensors, the test samples were clamped onto the load cell and tested by dropping the hammer from varying heights, as shown in Table 1. The peak forces of the impact appear to saturate past 80 degrees. Figure 3 displays trials testing at 50 degrees, which shows that the softer sample had a higher impact force, however this may be due to the oversaturation of the analog input from the load cell amplifier.
Further development of the impact test would involve testing on prosthetics and attachments; due to time constraints, only samples of stiff vs. soft materials were tested. Additional troubleshooting and modifications could be made to improve the impact test, such as a metal framework, a physical trigger to begin recording the impact, and an SD card attachment to log data, the alignment of the accelerometer closer to the axis of rotation, could all improve the overall performance of the apparatus.
