Content area
Full text
Ever had a motor zapped by a transient peak voltage?
YOU'VE SURELY HEARD ABOUT IT BY NOW"standing waves," something about the length of the cable run from inverter to motor.
Not only has the issue been aired in Electrical Apparatus (November 1996, "Are ASDs more trouble than they're worth?"), but in more than 20 articles in other publications, plus 50 technical papers presented before professional groups in several countries. Also, many electric motor manufacturers have issued explanatory literature.
So, what's being done about it? Can a motor user ever be sure of long motor life in the latest adjustable-speed drive applications?
The problem results from the operation of pulse width modulated (PWM) inverters, used almost universally today in ASDs involving low-voltage, polyphase a-c motors up to at least 500 hp. Some products range up to 900 hp. The newest PWM apparatus uses transistors in the output bridge. On-off switching of those devices involves frequencies of 5,000 to 20,000 Hz.
The good news is that the current waveform seen by the downstream motor is nearly a pure sine wave. The bad news is that the high frequency tends to make the circuit between inverter and motor behave like a radio antenna (depending upon the circuit length), developing "standing waves" or "ringing." Individual voltage pulses reflect on and reinforce one another, building up to 1,500- to 2,000-volt peaks at the motor terminals.
Those voltage peaks endanger any motor's 600-volt insulation system. In addition, the abrupt on-off transistor switching causes steep wavefronts to appear on the motor winding, such that much of the voltage pulse magnitude appears across lead-end coils. In a random winding, beginning and ending turns of any coil may touch each other so that the overvoltage appears not just across the coil, but between those two adjacent wires. This leads to winding failures that occur quickly and unexpectedly, but randomly. Figures 1 and 2 illustrate the typical mode of failure.
Taking the bad with the good
If that is characteristic of transistorized PWMs, why do they enjoy such widespread use? There are several reasons. One is that nice output current waveform. Earlier inverter designs caused their own motor problems-noise, added heating, loss of torque-arising from distorted inverter output. Another reason for new PWM...