While simulation can be helpful for the preliminary design and assessing feasibility, prototyping and bench testing is necessary for verifying actual performance. Simulation sometime does not account for all parameters and therefore can not exactly match real world test cases. Bench tests reveal some of the possible discrepancies from designed simulations, thereby exposing subtle neglected influences in the preliminary design phase and as well as offsets that can occur in the manufacturing and assembly process.

For assessing the hardware design, TMAG5273 measurements and joystick movement were captured simultaneously with the setup shown in . In this setup, the phone camera captures a video of the trigger moving while a computer screen captures measurements output in the TMAG5273EVM GUI. Both video files are then synced through an audible impulse. In the phone-captured video, a pointer is attached to the joystick shaft to help visualize joystick angle compared against a protractor centered at the hinge origin. To simplify assessment, movement was restricted by a 3d printed slot to allow only one joystick hinge to rotate at a time.

From the Figure 2-16 setup the measurements in Figure 2-17 were collected.

For assessing the software processing algorithm, a separate test setup was used. A 3D printer was outfitted with a special attachment to control the movement of the joystick as shown in Figure 2-18 and custom g-code was generated to direct the 3D printer how to move. From this test setup, the Figure 2-19 data in was collected.

The data presented in this section correspond to the final sets acquired after various iterations of debugging the test setup and adjusting for some assembly flaws. The prior collected data revealed potential issues that were not fully accounted for in the preliminary design calculations.