Induction motor speed controller an intriguing design exercise
This month we are very pleased to present the first article on an Induction Motor Speed Controller. For years we have had requests for just such a project but we have always had to decline on the grounds that we did not really have sufficient expertise in this area. In fact, I was probably being too conservative in this regard because we had already produced a similar project of greater complexity some 20 years ago in a 5-part series on a 2kW sinewave inverter, between October 1992 and February 1993. That project was quite advanced for its time and did not have the benefit of a fancy microcontroller. In fact, that sinewave inverter was produced at a time when such products, particularly of that high rating, were rare and very expensive.
By contrast, these days variable speed drives (VSDs) for induction motors are very common but they are still relatively expensive. Having said that, variable speed drives for single-phase induction motors are still not common. Now though, we have a design which will drive both 3-phase and single phase induction motors with name-plate ratings up to 1.5kW (2 HP). And while its most popular use will probably be to enable power savings with swimming pool pumps, it will also be of considerable interest for enthusiasts who wish to run lathes and other machinery over a wide speed range.
The whole concept of induction motor speed control is intriguing to anyone with an understanding of how these motors work. Tesla, the inventor of the induction motor, was definitely a genius and his concept of the rotating magnetic field (in a 3-phase motor) is such an elegant concept. So if you understand how an induction motor is effectively locked to the mains frequency, it becomes clear that simply reducing the voltage is not enough to vary the speed; the frequency must also be altered.
And while we have understood the concepts well enough, the idea of being able to control a 3-phase or single-phase 230VAC induction motor when you only have an input of 230V, and thereby 325V DC, is yet another hurdle. This design does it by a very clever technique. First, it makes the assumption that any 3-phase motor with a rating of less than about 2kW can be connected in delta configuration and this means it can be driven with 230VAC rather than 400VAC (equivalent to 415VAC with a 240VAC single-phase mains supply).
Second, the design manages to obtain a sinusoidal 230VAC between each phase. That is a big leap because how can you manage to obtain more than 115VAC (sinusoidal) by pulse-width modulation from a 325V DC bus? And if you only have 115VAC coming from each of the three phases, how can you possibly obtain sinusoidal 230VAC between phases when you can only expect 200VAC (ie, 1.732 x 115)? The answer turns out to be by not generating sinusoidal phase voltages. Instead, they are deliberately distorted (or squashed) by the addition of the third harmonic. The resulting inter-phase voltage IS sinusoidal and it is 230VAC.
Some people will regard this a technical skulduggery but it works beautifully. Not only that, but by setting a frequency range of 1Hz to 75Hz, the controller gives a speed range from very low to very fast; much faster than an induction motor can run at its nominal mains frequency of 50Hz. For example, an induction motor with a name-plate rating of 1440 RPM will have a theoretical speed range from less than 30 RPM to around 2160 RPM, albeit with reduced power if not running at 50Hz.
There is much more to the story, beginning on page 16 of this issue. I hope that you find it as intriguing as we have. Thanks to Andrew Levido for his clever design.