Electric motor power efficiency has taken center stage. Individuals, corporations, and governments are increasingly interested in saving power, now that technology can make it possible and economy demands it. Advances in motor control algorithms and cost-effective electronic components for implementing motor drives are creating a revolution in virtually every electric motor market. Control of the power factor in an efficient manner also means less lost energy, both in the motor and drive electronics, and in the power grids supplying the electricity to the homes, offices, and factories where the motors are used.

Potential Savings

The potential energy savings are staggering. Over 40 million electric motors are used in manufacturing operations in the United States alone.1 Electric motors account for 65 to 70 percent of industrial electrical energy consumption and approximately 57 percent of all electrical consumption worldwide.2 Saving even a few percent of the world's estimated 16,000-plus terawatt-hours (TWh) annual consumption of electricity amounts to several hundreds of trillions of watt-hours per year. Currently, the average motor in use today has an efficiency of 88 percent in converting electrical into mechanical energy. Figures on the order of 96 percent conversion efficiency are technically feasible for larger motors.

For comparison, the electrical generation capacity of photovoltaic solar cells in all of Europe, where both Germany and Spain currently lead the US in installed base, is projected to be only 15 TWh/yr in 2010. In the UK alone, with an annual total electrical consumption of approximately 350 TWh, the Institute of Engineering and Technology estimated that 5 TWh could be saved annually through the use of more efficient electric motors. Furthermore, many motors are not used in an efficient manner. For example, the motor may be oversized for the job at hand, or much of its mechanical output power may be wasted, meaning that additional savings may come from how the motor is used, on top of the savings from the motor itself. In 1996, the United States Department of Energy speculated on savings of 5 TWh per year by 2000, and a 100 TWh per year savings potential by 2010,6 considering both motor and related system-level savings.

The potential is there to make significant advances in the next few years as older motors and drives are replaced by newer more efficient ones. Because of the cost savings in electricity, many industries are voluntarily accelerating the turnover of their installed motor base, even replacing motors before they wear out. This is because the payback for the newer, more efficient motors and drives can be realized in less than a year and usually less than two years. Great strides are already being made. In the UK, for instance, sales of the least efficient motors, grade Eff3, have dropped from 68 to 8 percent between 1997 and 2004. During the same period, sales of the most efficient grade (Eff1) have increased from 2 to 7 percent, and further jumped to 17 percent in 2006, with the middle grade (Eff2) making up the balance of sales.

Regulatory influences on motor efficiency

Governments around the world are providing regulatory pressure to use more efficient motors. Starting with the Environmental Protection Act of 1992, which mandated motor efficiency standards and took effect in 1997, the United States government has been steadily increasing regulations. There are other voluntary incentives as well, such as National Electrical Manufacturers Association's (NEMA) Premium efficiency labeling standard (2001). Australia implemented standards on motors ranging from 0.73 KW to 185 KW in 2001 and tightened efficiency requirements in 2006. Very recently (March 2009), the European Union passed mandatory Minimum Efficiency Performance Standards (MEPS), which will be phased in from 2011 to 2017. Brazil (2002) and China (starting in 2010) also have current or planned mandatory standards. See the figure below for a comparison of efficiency requirements for various-sized motors in several jurisdictions, including the voluntary NEMA and Consortium for Energy Efficiency (CEE) standards, versus the wide range of efficiencies of available motors. The lines depict several mandatory and voluntary world-wide motor efficiency standards and the highlighted area represents commercially available motors.

Efficiency Range of 1800 RPM Motors Click on image to enlarge.

Motor controllers

Electric motor savings are achieved in several ways. The first is in the motor design itself, through the use of better materials, design, and construction. Another is by optimizing the mechanical angle between the various rotating magnetic fields inside the motor. This is done using the newer family of motor control algorithms, generally referred to together as space vector control, flux vector control, or field-oriented control. By keeping the magnetic fields of the rotor and stator oriented with the optimal angles between them under various speed and torque conditions (typically near 90 degrees), the motor can always be operated at peak efficiency. As a side benefit, other characteristics can also be optimized, such as fast and stable dynamic response to load changes, precise control of speed or torque, soft starting and braking, prevention of stalling at low speeds, high starting torques, and fault detection; often without sacrificing much in the way of overall energy efficiency. Some of these features were once obtainable only from a more expensive motor type, but can be achieved with the now ubiquitous, low-cost, and reliable AC induction motor, which comprises 90 percent of U.S. motor sales. One of the most significant advantages of the newer control algorithms is efficient variable speed operation.

A very large opportunity for system-level energy savings comes from using variable speed motor drives. A well-designed pump or fan motor running at half the speed consumes only one-eighth the energy compared to running at full speed. Many older pump and fan installations used fixed-speed motors connected directly to the power mains, and controlled the liquid or air flow using throttling valves or air dampers. The valves or dampers create a back pressure, reducing the flow, but at the expense of efficiency. This is probably how the HVAC forced-air system works in your office building; dampers control the airflow into each workspace while the central fan, which is sized for peak requirements, runs at full speed all the time - even if the combined airflow requirements of the building are currently very low. Replacing these motors with variable speed drives and eliminating or controlling the dampers more intelligently can save up to two-thirds their overall energy consumption.

Power factor

One often overlooked aspect of overall motor drive efficiency is the power factor. The power factor relates the shape of the current waveform drawn by a load to the sinusoidal voltage waveform supplied by the power company. If a load looks purely resistive, then the current drawn by the load is a sinusoid exactly in phase with the voltage waveform, and the power factor is unity. This is the most efficient condition.

If the load appears to be inductive, as it does in many motors, the current will lag behind the voltage in phase, and the power factor will be less than one, according to the cosine of the phase angle. Capacitive loads, which cause the current to lead the voltage, also reduce the power factor below one. In either case, the energy supplied to the motor will not be used optimally. Since the peak (and shape) of the current sine wave does not line up with the peak of the voltage sine wave, the instantaneous product of voltage times current averaged over a full cycle is lower. This is called the true power and is measured in watts.

The figure below is a vector diagram showing the relationship of apparent power to true (useful) power.

Power Triangle Click on image to enlarge.

Since the mains voltage is fixed, a higher current is required from the power company to compensate for the phase shift and deliver the same usable power to the motor, bringing the usable power (in watts) back up to the level required to do the desired mechanical work (in horsepower, for example). The product of this higher RMS current and the RMS voltage (measured in volt-amps) is called the apparent power. In many respects the power company has to build the infrastructure and pay for the higher apparent power, even though only the true power is doing useful work for the end user. This higher current means more losses as the power company generates and distributes the power. Power-line transformers can heat up and fail. Power losses go up as the square of the apparent power. A power factor of 0.7 means an apparent power of 1.4 times the true power, with nearly double the losses compared to a power factor of one.

Higher capacity circuit breakers will be needed on the branch circuit where the motor is used. Voltage drops on power distribution wiring will as much as double, necessitating an even higher current for the same delivered power. There are higher resistive losses in the motor as well, creating more heat and a shorter motor life. Alternatively, heavier wire must be used for the windings, reducing the number of turns and hence the efficiency of the motor. The reactive component of the current, which is out of phase with the voltage, is accomplishing no useful work, yet it creates additional losses in the overall system above and beyond those of the in-phase component that is doing all the real work. In cases where inductive or capacitive loads are linear, the power factor is often expressed as the true power divided by the apparent power.

Because of the extra capital and operating costs imposed upon them, it is very common in industrial settings for power companies to add surcharges for power factors below 0.95, though this is rarer in residential settings where the price of power reflects average residential power factors and the associated costs.

Why worry about power factor?

In an industrial setting, you can reduce your electrical bill by cutting power factor surcharges from the power company.

You will be able to put more true load on your branch circuit, since reduced reactive load currents will flow through your circuit-breaker junction box. Efficiency-sapping voltage drops in your branch circuits may also be reduced.

Finally, you might be required to worry about power factor by government regulations. European countries require power factor correction for power supplies rated over 75 W (IEC 555) and limit the harmonic distortion a power supply can inject into the mains though IEC/EN61000-3-2. These regulations require controlling the input current distortion up to the 40th harmonic of the line frequency.

Combating low power factor

Fixed-speed AC induction motors connected directly to the mains voltage look primarily inductive from the point of view of the power plant and distribution grid. To combat the inefficiency this causes (and the corresponding surcharges from the power company), industrial concerns will often add a compensating capacitive load to the power line. This shifts the phase of the power line current so it is back in phase with the voltage. Since the added capacitive load is mainly reactive, it dissipates almost no power itself, except due to non-idealities such as non-zero series resistance and leakage.

Fixed-value capacitors can be applied or removed automatically by a centralized power factor controller, based upon measurements of reactive currents as factory motors are turned on and off. Another scheme is to use an unloaded motor-generator as a sort of synthetic capacitor called a synchronous condenser; usually one such machine for a whole factory full of motors. The closer the compensation capacitors are to the motor(s), the better, as there are still reactive currents flowing back and forth between the inductive and capacitive reactive loads. Note that the current component supplied by the power company can be made to look almost purely resistive with the right compensation, localizing the reactive part of the load current so it does not have to go over the long transmission lines from the power company to the factory - and this also keeps it off your electric bill.

Power Factor Correction for modern motor drives

While modern motor drives provide many features and can greatly enhance the efficiency of the motor itself, and can often make the whole system containing the motor more efficient with new features such as variable speed control, there is at least one downside of the new technology. Everything you have learned about correcting the inductive power factor of electric motors using capacitors is irrelevant.

The bad news is that simple capacitive power factor correction schemes used in the past do not work well for modern motor drives. With the new generation of motor controllers, the motor drive electronics look like a large AC-to-DC power supply when viewed from the power grid. Without power factor correction, these look highly non-linear. A quick look at a motor drive block diagram reveals why. Motor drive electronics usually consists of two main parts: a rectifier that converts the AC input mains voltage to an intermediate DC power bus and an inverter that converts the DC bus voltage to AC at the motor's operating frequency and current. In many ways, these two main blocks are duals: one efficiently converts AC-to-DC, and the other efficiently converts DC-to-AC. The energy losses in these two blocks, though they may seem a roundabout way to go from AC to AC, are more than offset by the efficiencies gained in having more control over the magnetic field phase and the added advantage of variable speed operation.

The figure below shows a modern motor drive, consisting of an AC-to-DC rectifier followed by a DC-to-AC inverter.

Modern Motor Drive Click on image to enlarge.

Nearly all single-phase AC-to-DC power supplies have a full-wave bridge rectifier circuit on the input, followed by a large bulk capacitor, which attempts to hold its DC voltage constant between the half-cycle peaks of the input voltage sine wave. Of course, no matter how large it is, the capacitor droops slightly between half-cycles, so when the next peak comes, the rectifier bridge conducts and recharges the capacitor. The capacitor charging current only flows when the input voltage (less the voltage drops across the rectifiers) is greater than the voltage on the capacitor; when it is less, the rectifiers are off and little or no current flows. Therefore, the current is highly non-sinusoidal, as shown in the figure below. The low power factor caused by the high harmonic content of the currents causes problems for the power company that are similar to those caused by sinusoidal reactive power—only worse. The harmonics cause distortion in the voltage waveform, and can even cause destructive resonances in the power grid.

The figure below shows that a simple rectifier without power factor correction (PFC) draws current from the AC mains with a high harmonic content, and hence a low power factor.

Simple Rectifier without Power Factor Correction Click on image to enlarge.

Uncorrected power factors may be as low as 0.5 or 0.6 for this type of rectifier design. A similar situation applies to three-phase mains power, but the rectifier bridge has six diodes instead of four, and the phase peaks six times per cycle instead of twice.

For lower power systems (<100 W), passive power factor correction (PFC) can be used. For these low-power applications, the energy efficiency of passive PFC can be relatively high (for example, 96 percent efficiency). A low-pass filter, usually comprised of an inductor, capacitor, and resistor, is inserted between the AC mains input and the bridge rectifier. This tends to draw most of the current out of the mains at the line frequency, which is within the passband of the filter. The harmonics of the line frequency are reducedto some degree, and the current waveform is smoothed out. Passive power correction is generally not sufficient for motor control applications due to its marginal performance with respect to the resulting power factor (typically around 0.75), and the large size of the components for higher power applications.