Characterization of Sensorless Motor Control Technology Using solidThinking Embed

• Analytics & Modeling - Digital Twin / Simulation
• Processors & Edge Intelligence - Embedded & Edge Computers

• Automotive
• Electronics

• Product Research & Development

• Digital Twin
• Virtual Reality

• System Integration
• Testing & Certification

Texas Instruments (TI) was faced with the challenge of characterizing their FAST™ observer, a part of their InstaSPIN™ technology. This technology enables designers to identify, tune, and fully control any type of three-phase, variable speed, sensorless, synchronous or asynchronous motor control system. The task was assigned to Dave Wilson, Senior Motor Systems Engineer with the C2000 group. Wilson attempted to characterize the FAST™ observer by setting up a dynamometer (dyno) system with a circuit board to control it. However, this process was slow, tedious, and required constant recalibration due to output variances over time and temperature changes. Furthermore, the electromagnetic torque could not be measured on the dyno, only the shaft torque could. This was a problem since the software could not be properly tested as the hardware he was using was not adequately equipped to test it.

The customer in this case study is Texas Instruments (TI), a renowned American technology company that designs and manufactures semiconductors and various integrated circuits, which it sells to electronics designers and manufacturers globally. TI is one of the top ten semiconductor companies worldwide, based on sales volume. TI's products are used in a wide range of applications, including digital signal processing and analog technologies, embedded processors, and software development tools. In this particular case, the focus is on TI’s InstaSPIN™ technology, a motor control technology that enables designers to identify, tune, and fully control any type of three-phase, variable speed, sensorless, synchronous or asynchronous motor control system.

To overcome the challenge, Wilson turned to solidThinking Embed, a tool he had become familiar with four years prior. Embed provided him with a solution to create fast and accurate simulations of motor analog dynamics as well as the digital control. He was then able to automatically create C code from the controller portion of his graphical diagram, and download the code to run on the Piccolo target. Using the Embed JTAG Hotlink in a new synchronous mode, Wilson could run the motor simulation in lock step with the control running on the target in non real-time. This allowed him to verify controller operation against any motor configuration, no matter how big or small. It also allowed him to take part of the simulation and interface it to the FAST™ observer. He developed an Embed simulation of the entire system except for the FAST™ piece of code. He could now control any gains he liked. Different parameters like resistances, inductances, controller gains and voltage tolerances could now be controlled as well.

• The use of solidThinking Embed in the simulation process resulted in significant operational benefits. It completely solved the problem that TI was facing in characterizing the FAST™ observer. The solution allowed the generation of the FAST™ TRM (Technical Reference Manual), and also resulted in a very useful tool. This tool can enable TI’s customers to test their motor control requirements using TI’s actual ROM based algorithm without having to hook up an actual motor system. This gives them a quick way to test the InstaSPIN™ FOC (Field Oriented Control) without having to build a whole system for it. The speed of how its drivers work and being able to interface directly with TI's ROM code was a key enabler for the project.