Content area
Full Text
Download this article in .PDF format
This file type includes high-resolution graphics and schematics when applicable.
All position-feedback devices for motor-driven machines have drawbacks. Encoders are usually rugged or inexpensive, but not both, and often necessitate expensive housings for industrial-level protection. Rotary encoders (optical and magnetic) have resolutions limited by the diameter of the disc and the width of the lines involved. Traditional resolvers are no better, because they need a dedicated excitation driver plus a hardware or software-based resolver-to-digital (R/D) converter (in addition to the resolver itself). They're also "blind" for a portion of the electrical cycle - when drive voltage is near zero and sine and cosine winding outputs are at zero. Therefore, to feel through blind periods and estimate rotor position, resolvers use tracking R/D converters, which can degrade the phase-margin. Dual-driver resolvers address this problem but add cost for the drive electronics. In fact, for small motors, many feedback devices cost more than the motor itself. Options that don't, such as Hall sensors, have temperature and resolution limitations.
Now, another position-feedback option in a combination motor-and-resolver design eliminates these problems with minimal materials and electronics. For the purposes of this article, we call the resulting feedback an emulated resolver, as the setup acts as a traditional resolver but without the rotary transformer or separate driver electronics. These sensors track motor position in two steps:
1. Emulated resolvers tap into the drive's pulse-width modulation (PWM), normally used only to control the motor. Using sample windows that are synchronized to the PWM timing, they generate resolver-type sine and cosine signals.
2. Emulated resolvers measure the changing magnetic paths that link the motor's rotor and stator. To do this, they use dedicated sensor coils tucked into open space inside the motor. Motors that physically accommodate the sensor coils have lateral slots in the stator to let the coil wires pass across the faces of each pole structure without mechanical interference with the rotor. The sensor coils consist of a few turns of wire so the sensors cost less and are more compact than separate encoders or resolvers. The coils integrate into the motor, so they can't become misaligned like other feedback elements - and the commutation signals are also inherently phased. Plus, the coils...