Automobile traction control and anti-lock braking systems rely on a real-time approximation of the torque distributed through each wheel. Engine mapping, the current method of torque approximation in automobiles, generally loses accuracy over time, reducing the efficiency of the system and increasing the likelihood of safety hazards while driving.
Over the fall and winter terms of my senior year in Thayer, I collaborated with a project group to design and build an optical torque sensor for automotive applications. The project was sponsored by Analog Devices Incorporated, a Massachusettes-based semiconductor manufacturer. The goal of the project was to use two ADI photodiodes to measure the angular deflection of a rotating car axle and calculate the torque in the axle in real time based on its material properties.
The project's main deliverable was a mechanical test rig, scaled to model typical automotive drive trains. Analog Devices will be able to use the rig to characterize current, as well as future photodiode technology in order to improve their torque sensors. All rig components were designed in SolidWorks 2016 and fabricated in the Thayer Machine Shop.
My main contributions to the project included determining specifications for the rig, ordering stock metals, machining rig components, and characterizing the motor controller. Sensor designs and photodiode characterization was managed by two electrical engineers in the project group.
The mechanical specifications for the test rig were determined through a parametric study which related drivetrain characteristics of ten common car models. The group was able to gain access to at least one of each car model, and all of the data in the study was either measured directly from the sample vehicles or taken from manufacturers records specific to each model of car.
Once the rig was fully assembled, a control circuit with an LCD monitor was added to display the rig's rotational speed and brake line pressure in real time for the operator. The control circuit was driven with an Arduino Uno microcontroller.