Learnings:

• Team and resource management.

• Engineering modelling and optimization of components.

• Systems Engineering based approach to design.

• Design for Manufacturability and Assembly.

• E Motor design and testing fundamentals.

Overview:

As a member of Purdue Electric Racing (University SAE Team) I lead the design of a flywheel inertia dynamometer to test 15kW motors from 2022 - 2023. The work showcased below includes work by other members of the team who worked with me on this project.

Problem:

• Limited Access to motor testing facilities on campus.

• Detailed Motor characterization needed for torque vectoring design.

• Controlled test bench needed to validate motor cooling performance.

• Power electronics and battery tests needed under load.

Solution:

Primary Research and Idea Generation:

In summer 2022 a study into motor dynamometers was performed to understand the different ways we could execute the design and which type of test system would work with our budget and resources. We spent 2 months in the summer attempting to design a motor dyno using an existing Emrax motor as the absorber. Through the design phase we learned a lot about the fundamentals in motor testing. The motor dyno design had to be scrapped due to the inability to accurately measure applied torque within our budget. The team decided to shift to an inertia flywheel dyno to achieve our goals.

Problem Definition:

For the inertia flywheel dyno, the following primary design requirements were created:

• Capable of characterizing the torque rpm curve of a Plettenberg Nova 15 motor.

• Maximum torque rating of 25Nm.

• Max. Operating RPM of 12000.

• Total cost under $2000.

• Dyno assembly movable by 1 person.

• Closed-loop operation without need for external stops.

• Safe to use and operate within 2 feet of the device.

• Design should be easily upgradable to accommodate future motor and gearbox improvements.

These design metrics helped us define and limit our design scope.

Flywheel Design:

The inertia flywheel is one of the most important subsystems within this design project. It is the known well-defined load within our system that can be used to characterize torque performance of the motor. The calculation of the required load and the dimensions of this flywheel were done through a modelling based approach. The RPM at the max. torque was used to identify the required rotational inertia of the sytem. A challenge with the inertia was optimizing it be made with a large enough radius to reduce mass, while also ensuring the flywheel could be fabricated with the equipment available to us. The energy of the flywheel also determined its Factor of Safety and the braking torque needed to slow it down. A basic analytical model of all these parameters helped identify the optimal radius, mass, length while also ensuring that a brake system could slow the wheel down safely.

The flywheel is made of three steel plates waterjet cut of steel sheet. The three plates can be removed to reduce the inertia of the system. The entire flywheel is balanced by external professionals before being mounted on the shaft using a shaft coupler. The image below is a CAD of the flywheel.

Frame Design:

The frame for the structure was made primarily with Aluminum 8020 extrusion pieces. These were strong and easy to connect with several mounting locations for various peripheral components. The motor mount was used to secure the DUT to the Dyno and was designed to hold the torsional loads and weight of the motor. All components are bolted directly to the frame through different T-slot or 8020 attachments. A safety cage made of 1 in wood and 1/8th in steel sheet was created to surround the Flywheel. This was designed to catch any stray bolts or projectiles from the high speed dyno and protect the users. This safety cage can be taken off easily to access the flywheel when not in use. Several mounting sheets are there around the dyno for different peripherals and components.

Braking System:

A standard hydraulic disc brake system was used to slow down the flywheel motion after testing is completed or during an emergency. The brake cylinders are connected to normally closed pneumatic cylinders that are controlled by external valve system. When a brake command is given the pneumatic cylinders are released activating the brake system and slowing down the flywheel in under 3 seconds. The normally closed system ensures that in case of pressure loss or other issues with air supply, the brakes automatically engage and stop the system.

Electronics and Control System:

Controlling the dyno and collecting data from the system requires the use of an HV loop to power and control the motor and an LV loop (5V, 12V, 24V) to manage auxiliary systems and collect data from the system. The HV loop was a 300V 5KW power supply that interfaced directly with the motor controller. This power supply had its own internal flyback protection and voltage regulation along with E-stops in case of emergency. The LV system was controlled by an Arduino MEGA which acted as the central controller for the entire Dyno system. It interfaced with other LV components such as motor cooling pump, motor controller, pneumatic brake cylinders, etc. For data acquisition on this dyno we use a Hall Effect Sensor to record shaft RPM. We also have temperature probes in the gear box and motor cooling loop. All the components receive power from an external supply rail for 5V, 12V and 24V components. The monitoring of all relevant system parameters also allows closed loop operation.