SBIR Phase I: Low-Cost Ultra-Efficient 50-gm, 300-W Servoelectronics Module with Integral Sensors

Barrett Technology Inc, Newton MA

Investigators

Abstract

This Small Business Innovation Research (SBIR) Phase I research project will evaluate the feasibility of reducing manufacturing cost by an order of magnitude of a power efficient ultra-miniature, brushless servo-electronics module. The module integrates rotor-position sensing for high-performance servomotor applications. Starting with a recently proved prototype driving high-performance robotic arms, the innovation hinges on substituting a $500 ultra-precise 40,000-counts/revolution, optical encoding chip with a new $8 magnetic-encoder chip (array of Hall Effect sensors). Since the array is to be literally embedded into the servo-electronics, the technical challenge is to overcome parasitic variations in magnetic field by leveraging the existing 32-bit Digital Signal Processor (DSP) used for commutation. The research will identify and measure the effects of electromagnetic fields on the magnet encoder, validate a shielding-plus-DSP-compensated approach to reduce these effects, and measure the degree of precision drop and its effect on commutation efficiency and overall performance. As machines become more intelligent through embedded processing and sensor fusion we expect them to do more too, improving not only industrial productivity, but our quality of life. While embedded processors and MEMS-based sensors have become tiny, highly effective, and affordable, similar improvements in servomotors have evolved more slowly. At fractional-horsepower levels the power electronics contribute significantly to total motor-system bulk. By making brushless motors (including drive electronics) smaller and more efficient, the resulting devices and machines will become simplified, lighter, less obtrusive, and far more capable. Robots will become more agile with additional degrees of freedom and less mass to accelerate.