PROBING MYSTERIES OF THE INVERTER: PART I

Article Tools

Image Gallery

Print

E-mail

Comments

RSS

An inverter usually performs best when it's fully loaded, or close to it (see Fig. 4). For example, if you have a 500-watt unit, it will work most efficiently when you're drawing 400 to 500 watts through it. As the load drops, so does the mechanism's efficiency . . . since an inverter requires a certain amount of what's referred to as "tare" energy to keep it running. At low demand, operating power can become a large part of the overall output. In fact, this standby (or idle) current will be between 5 and 20% of the maximum power the device is capable of supplying, depending on the particular inverter.

The obvious solution to tare loss (which can drain your batteries of valuable energy) is to shut down the inverter when no power is being demanded. Several models on the market incorporate circuits that sense power demand and switch the inverter off when there's no load.

MODIFIED WAVEFORMS

You may have noticed that I used the word "mimics" when describing an invert er's attempt at producing AC power. While it is true that the simplified circuit shown in Fig. 2 would produce alternating current, the form of the wave generated by the two transistors would be far from sinusoidal ... which is the pattern of pure AC utility power, as illustrated in Fig. 1. In fact, the waveform from our simple inverter would be more properly called square, because of its characteristic shape (which is shown—interposed with a sine wave, for comparison—in Fig. 5A).

Now some AC equipment can handle squarewave AC power just fine, but other devices have real problems with it. Induction motors, for example, are designed to work on the purest type of AC current—the sine wave—and when the transistors in an inverter deliver a square wave (which represents at a minimum, a distortion of 40% from the sine wave) to one of these devices, the motor must work harder to overcome the counterproductive currents. The result is a marked increase in operating temperature and a loss of efficiency.

The ultimate solution, of course, would be to generate a sine wave. Unfortunately, this is easier said than done, since inverters capable of producing such a form are inherently bulky, expensive, and inefficient. A simpler remedy is to modify a square wave to remove as much of the distortion as possible. For example, if we delay the activation of the second transistor, we'll get a waveform that looks much like the one shown in Fig. 5B. As you can see, this modified square wave is an improvement. The distortion has been reduced from 40% to less than 20%.

RSS | Comments | E-mail | Print