It’s been a while that I want to write a post to discuss about the real impact of the electronic/semiconductor device selection on the energy consumption of electric motor-driven systems. It seems that every chip on the market has capabilities for “more energy-efficient” motor control but I never saw a single article actually showing that. The real question: is this possible?
As a first step, for those interested in the subject of power savings or energy-efficient motor drive systems, I highly recommend to read the latest report of the Internation Energy Agency called “Energy-Efficiency Policies for Electric Motor-driven Systems” published last year (2011). Like any similar report published in the last 20 years on this subject (like this one , this one or this one) the recommandations are :
1- Make sure your motor sizing (nominal power) is correct for your application. In large industrial motor applications this can have a very significant impact on the power consumption. The rationale is easy to understand, why would you drive a Hummer when the only thing you need is a Toyota (keeping in mind that the best way to save energy is to keep them in the garage) ? Correct motor sizing has nothing to do with the device or even motor control driver selection.
2- Use Variable-Frequency Drives (VFD). That’s the famous 20-50% power savings we see everywhere. Knowing that power output is the product of the torque and the speed, instead of fixing motor speed at nominal speed and varying the power output by varying the torque only, the idea of VFD is to regulate power using both speed and torque (hence the name variable-frequency).
It may be seducing to think that since the VFD motor control software is running on an electronic device, then it MUST certainly have an impact on the performance but it is not true. It’s all in the software and any energy-saving motor control software feature can be implemented on any kind of electronics device, wheter it is a MCU, a DSP or a FPGA:
– VFD via scalar or vector control
– flux optimization of induction motor and permanent magnet synchronous motors
– power switching and time-harmonic losses via advanced PWM algorithm (like this excellent recent article from Xilinx)
Why ? Because you need to think that electric motor are electro-mechanical devices and even though you think those system have “fast dynamics” they are still very slow (kHz range) compared to the computing bandwidth of most electronic devices on the market (MHz range). And even though you crunch number infinitely fast, you will always be limited by your power converter PWM frequency to keep switching losses in the power converter at a low level.
OK. But what about spindle drives running at 30,000 RPM ? Yes, those application needs high control bandwidth but they are not “power applications” since the motor that are typically used are low-power motors but mostly because the motor is not used as an electro-mechanical power converter (focus is on transfering energy) but rather as an eletro-mechanical actuator (focus is on the precise dynamics).
Bottom line: Device selection has no impact on electric motor power savings. It’s all about the correct motor sizing for the load profile, the correct power converter sizing and the motor control software (optmized for the application itself). Are there devices where this software is easier to develop and integrate ? That’s another question that relies more on the design tools and business model of the device manufacturer than the device itself.