Looking at electric drive efficiency

variable-frequency drives (VFDs) provide dramatic energy savings in myriad applications with varying loads and speeds. While power and energy savings are key benefits, VFD efficiency is also of interest because drives can experience significant efficiency losses at partial load operation-much like electric motors.

07/01/2009


Five makes of VFDs tested had efficiency variations (bands) as shown for 50 hp (37 kW) units running at different speeds - along with larger characteristic efficiency drop-off at reduced torque and speed.

Few will dispute that variable-frequency drives (VFDs) provide dramatic energy savings in myriad applications with varying loads and speeds. While power and energy savings are key benefits, VFD efficiency is also of interest because drives can experience significant efficiency losses at partial load operation—much like electric motors. Yet motors have received predominant attention for energy efficiency, including development of recognized test procedures and standards. A broader look at “motor system” efficiency must recognize the effect of VFDs.

Recent progress in this area includes work done at the energy technology labs of Hydro-Quebec's (Canada) Research Institute. Findings of the underlying study, “Variable Frequency Drive Testing,” have been presented at Motor Summit 2008 (Zurich, Switzerland), MEPSA 2009 (Sydney, Australia), and 2009 Motor, Drive & Automation Systems Conference (Orlando, FL) by Pierre Angers, an engineer at Hydro-Quebec.

“Published VFD efficiencies at other than full rated power are scarce,” says Angers. “There is no recognized VFD efficiency standard for comparing different makes of drives.” Efficiency is the highest around 100% speed, which is just where you don't want to use variable-speed drives, hence most available efficiency values are virtually “useless,” Angers explains.

Goals of the project were to establish an efficiency testing procedure for VFDs, motors, and VFD-motor combinations—as well as to enable efficiency comparison between various VFD systems. Angers has tested five major makes of VFDs, which powered three sizes of induction motors (10, 50, and 100 hp/7.5, 37, 75 kW) at various loads and speeds. For each test run a series of 20 test points ranged widely over “% speed” and “% nominal torque” values (25-100% speed and 10-100% torque). Drive manufacturers' names remain proprietary. Test setup measured efficiency as a ratio of output to input power.

Preliminary results show 2-8% variation in VFD system efficiency, especially at low loads and speeds. Efficiency variation between different brands of VFDs was substantially less (see diagram). However, variability of VFD efficiency is further impacted by factors such as drive temperature, supply voltage (and voltage unbalance), switching frequency, and drive options like active front-end, reactors, or output filters.

One novel aspect of Angers' research is presentation of motor system efficiency results as a contour map, with percent nominal speed and percent torque serving as other variables. A typical conveyor application plots as a horizontal line (constant torque) across the efficiency contours, while a pump/fan application plots as a cubic system curve.

The work described here is the basis of a protocol for testing VFD efficiency. Two proposed standards are in draft status: Canadian standard (CSA C838) Variable Frequency Drives and International Electrotechnical Commission standard IEC 60034-2-3.

Users of VFDs obtain energy savings from matching motor speed to load requirements; in addition, they need to be able to compare efficiency of different electric drive brands for their applications.


Author Information

Frank J. Bartos, P.E., is a Control Engineering consulting editor. Reach him at braunbart@sbcglobal.net .




No comments