ASHRAE AB-10-024-2010 Heat Gain from Adjustable Speed (Variable Frequency) Drives.pdf

1、608 ASHRAE TransactionsThis paper is based on findings resulting from ASHRAE Research Project RP-1395.ABSTRACTThe objectives of this paper are to provide information on heat loss for solid state VFDs at both full and part-loads and to show how these losses could be predicted. In developing these pre

2、dictions, use is made of manufacturer and independently measured data. Two analytical part-load VFD power loss models are presented, a linear model in which the VFD power losses vary linearly with the load and a quadratic model in which the VFD power losses vary as the square of the load. The perfor

3、-mance of these two analytical models is compared with measured part-load VFD power loss data. An example of VFD power loss at part-load is presented and compared with the results of the two models. The performance of the VFD power loss quadratic model is shown to be superior to the linear model.INT

4、RODUCTIONThe Adjustable Speed Drive (ASD) is also known as an Adjustable Frequency Drive (AFD) or Variable Frequency Drive (VFD). ASDs are used in industrial processes that need to control the shaft speed in rotation equipment such as fans, pumps and assembly lines. ASDs can be mechanical where a tr

5、ansmission with belt or chain could be used, hydraulic, elec-trical where resistors and special motors could be used, and electronic using solid state components. AC motors are used more frequently than DC motors because they can be connected directly to the utility power lines. For this reason, VFD

6、s have greater use than other ASDs. Electronic ASDs or VFDs are popular because they are easy to use and are energy efficient. Efficiency, functionality, and the economics of VFDs have motivated increased use of these devices.The VFD is one type of ASD used for AC motors. Here-after, VFD will be con

7、sistently used. VFDs are used in indus-trial processes that need to control the shaft speed of AC motors in assembly lines, fans, pumps and other mechanical equipments.The VFD can be diagrammatically represented by three blocks: an AC-DC rectifier, a DC bus link and a DC-AC inverter as shown in Figu

8、re 1. The principle of the VFD is to transform the stable 50 or 60 Hz power line frequency into a variable frequency through two steps. First, the sinusoidal input voltage of the rectifier is converted to a DC voltage. Following the rectification, the DC voltage is chopped into an AC output voltage

9、of the desired frequency according to the required AC motor speed. While the input frequency of the AC-DC rectifier is always constant, the output frequency of the DC-AC inverter is variable causing modification of the AC motor speed.The efficiency of a VFD is related to its power loss. The VFD powe

10、r losses consist of both constant and variable quan-tities. The constant losses are comprised of cooling fans and electric circuit excitation and these losses do not depend on the load. The variable losses consist of switching and lead losses in the power semiconductors and these losses do depend on

11、 the load as pointed out by Lidenborg (1998). VFD power losses at part-loads are needed by HVAC engineers and designers for correctly estimating the indoor environmental heat gain. Two analytical VFD power loss models will be presented and compared in later sections. These models consist of a linear

12、 model in which power losses vary linearly with the load and a quadratic model in which the power losses do not vary linearly with the VFD load. Heat Gain from Adjustable Speed (Variable Frequency) DrivesEmilio C. Piesciorovsky Warren N. White, PhDE.C. Piesciorovsky is a graduate student in the Depa

13、rtment of Electrical and Computer Engineering and W.N. White is an associate professor in the Department of Mechanical and Nuclear Engineering at Kansas State University, Manhattan, KS. AB-10-024 (RP-1395)2010, American Society of Heating, Refrigerating and Air-Conditioning Engineers, Inc. (www.ashr

14、ae.org). Published in ASHRAE Transactions (2010, Vol. 116, Part 2). For personal use only. Additional reproduction, distribution, or transmission in either print or digital form is not permitted without ASHRAEs prior written permission.2010 ASHRAE 609The organization of this paper consists of (1) th

15、e review and update of full load VFD power losses determined by manufacturer data, (2) the review of the linear part-load loss model, (3) the presentation and use of measured part-load power loss data from two sources to produce the quadratic part-load loss model, (4) the verification of the quadrat

16、ic model by comparing the predicted results with part-load power loss measurements from a third source, and finally (5) an example showing the use of the quadratic model in deter-mining the heat gain of a partially loaded VFD.ANALYSISVFD Full Load Power LossPower losses in VFDs at full load vary lin

17、early as a func-tion of rated power. This is true for line to line voltages of 240, 480, and 600 volts. In this study, VFD power loss data from four manufacturers were collected. The number of VFD units considered exceeded 150. The data were grouped according to voltage and rated power. For 480 volt

18、 and 600 volt VFDs, the units were further divided into units having power ratings below 50 hp and those units having a power rating of 50 hp or higher. Figures 2 through 6 show the VFD power losses as a function of rated load. Each figure shows a collection of manu-facturer data together with a lin

19、ear regression of the accumu-lated data. Also shown in Figures 2, 4, and 6 is a comparison of the information presented by McDonald and Hickok (1985) and the linear regression presented by White, Pahwa, and Cruz (2004) where VFDs having power ratings of only 50 to 500 hp were considered. The differe

20、nces between MCDonald and Hickok and the other results probably stems from the VFD technology in use at the publication time for their paper.Figures 3 and 5 include VFDs having ratings below 50 hp for 480 and 600 volt units, respectively. Figures 4 and 6 include VFDs having ratings of 50 hp or great

21、er for 480 and 600 volt Figure 1 Variable frequency drive components.Figure 2 240 volt VFD power loss.Figure 3 480 volt VFD power loss for units rated up to 50 Hp. Figure 4 480 volt VFD power loss for units rated from 50 to670 Hp.2010, American Society of Heating, Refrigerating and Air-Conditioning

22、Engineers, Inc. (www.ashrae.org). Published in ASHRAE Transactions (2010, Vol. 116, Part 2). For personal use only. Additional reproduction, distribution, or transmission in either print or digital form is not permitted without ASHRAEs prior written permission.610 ASHRAE Transactionsunits, respectiv

23、ely. Figure 2, 4, and 6 show small differences between the data collected in this work and those accumulated by White et al. (2004).The regression lines shown in Figures 2 through 6 are presented here. The power loss for 240 volt VFD units rated up to 100 hp is(1)where Plossis the VFD power loss at

24、full load in watts and Hp is the VFD rated power in horsepower. The power loss for 480 volt VFD units rated up to 50 hp is(2)For 480 volt VFD units rated from 50 to 670 hp, the power loss is (3)The power loss for 600 volt VFD units rated up to 50 hp is(4)For 600 volt VFD units rated from 50 to 670 h

25、p, the power loss is(5)Equations (1) through (5) are only valid to calculate the VFD power losses at full load. For loads less than the rated power, another means must be used to determine the dissipated power. This subject is addressed in the following section. VDF Power Loss at Part-Load White, Pa

26、hwa it probably doesnt include variable motor speed, the motor running from a pulse-width modula-tion (PWM) power source, or variable loads. Directly measur-ing electrical power from the drive is also difficult due to the PWM waveform of the voltage.Emilio C. Piesciorovsky: It is true that manufactu

27、rers report power losses for VFDs at rated power loads and do not include how the power losses of the VFD change for different loads. As a result, we compared measured data to that produced by our model. This measured data was reported by Chvala, Winiarski, and Mulkerin (2002) in Appendix A of their

28、 paper. The measured data were separated into the input power of the VFD and input power of the motor. Since the input power of the motor is equal to the output power of the VFD, the power loss of the VFD can be estimated as the difference between the input power of the VFD and the input power of th

29、e motor. Informa-tion about how the measurements of the motor and VFD were made are not explained in great detail by Chvala, et al., but these measurements were made at the Motor Systems Resource Facility at Oregon State University.2010, American Society of Heating, Refrigerating and Air-Conditioning Engineers, Inc. (www.ashrae.org). Published in ASHRAE Transactions (2010, Vol. 116, Part 2). For personal use only. Additional reproduction, distribution, or transmission in either print or digital form is not permitted without ASHRAEs prior written permission.