Power supply design determines an appliance’s energy profile
On the plane coming home from the Institute of Electrical and Electronics Engineers Power Engineering Society meeting, I was proofreading a paper that proposed to convert our way of looking at power-quality phenomena from sags, swells, transients and harmonics to a single unified parameter, energy. It was sort of like the attempts of Einstein and other physicists to find the grand unifying theory that can define the forces of nature, electromagnetic, gravity and weak and strong nuclear forces in terms of a single theory or relationship.
Though the paper wasn’t fully convincing that one parameter could describe all PQ, it did highlight the basic premise that it is all about getting the right energy at the right time to do the job. The corollary to that is the wrong amount of energy at the wrong time can result in equipment malfunction or even destruction.
The design of the power supply or the manner in which the equipment utilizes the electric power determines much of this energy profile. A simple resistive load uses energy in linear manner—the more voltage that it is given, the more current it will draw. And the current will change instantly when the voltage changes. It has no ability to store energy, so it only uses what it has, when it has it.
In most applications, devices that contain resistive loads (such as heating elements found in electric toasters, stoves or coffee pots) tend to be immune to most types of power quality phenomena. A transient would normally do little damage and provide no significant change in the temperature of the heating element (which is its job). This is also the case for most short-duration sags or swells. The thermal mass of the heating element is too big in comparison to the energy in the transient to have much of an effect. If the RMS value of the voltage waveform with lots of harmonics is the same as the RMS value of a waveform with no distortion, then the heat generated from both will be the same, so it can carry its job without regards to such.
This immunity enjoyed by resistive type loads is not the case for inductive or capacitive type loads. The impedance of inductive loads (such as transformers and inductors) is a function of the frequency. As the frequency goes up, so does the impedance. Harmonics will therefore cause the impedance to change, which even with the same rms voltage will now draw different amounts of current. The losses within such loads themselves usually increase as the square of the increase in frequency—the 11th harmonic as compared to the 3rd harmonic will cause losses that are 11 ¥ 11 divided by 3 ¥ 3 or more than 13 times greater, even though the frequency wasn’t even four times higher.
As inductive loads don’t like to change the current instantly, the capacitive loads don’t like to change the voltage instantly. That is why capacitors are used to “hold up” the voltage in power supplies. One could think of them as energy-storage devices, providing the energy when the voltage supply dips. There is a delay from the time the voltage changes to when the current eventually changes, just like the inductor had, but in the opposite manner—current leads the voltage whereas the current lags the voltage in an inductor. The capacitor’s impedance is also a factor of frequency, but in the opposite manner again as an inductor. As the frequency goes up, the impedance goes down.
The typical electrical-distribution system combines all three of these elements. The electricity supplier often uses capacitors to try to offset the inductance of the loads, so that the displacement power factor, or the angle between the voltage and current wave forms, is brought back to zero.
However, if the frequency of the harmonic currents present in the system is such that the impedance of the inductors and the impedance of the capacitors is a match, then a condition of resonance occurs and large damaging overvoltage conditions can occur. It is sort of like a tug-of-war where one side pulls the other toward the middle and then the second side rallies and pulls the first side back to the middle. The overvoltage can exceed the ratings of other equipment on the system. In addition, the harmonics can cause such high losses in the capacitors as to cause them to overheat and even go pop.
The so-called “sensitive” electronic equipment usually has an AC-to-DC rectifier made of either diodes or SCRs that conduct electricity only in one direction. This means there is one set that conducts when the voltage sine wave is positive, and another set that conducts when it goes negative. The current from each of these paths is used to charge up a capacitor, the energy-storage device. Another rule they follow is that they only conduct when the voltage on the anode side of the device is greater than the voltage on the cathode side of the rectifier.
Since the anode side is usually the AC supply, and the cathode is connected to the storage capacitor (which likes to keep its voltage steady), the current is only drawn in the middle of each half of the sine wave, when the AC voltage peak is greater than the DC storage voltage. If very little energy is drawn out of the storage cap, then the width of the current pulse is very narrow, as only a little energy needs to be replaced in the cap. As the energy drain increases, the width increases.
However, this results in a current waveform that is not a pure sinusoidal waveform, rather, one rich in harmonics. In addition, the remaining voltage during the sag even at the peaks of the waveform may be less than the voltage of the capacitors. So then no current will be drawn, as can be seen in the figure below. The load will keep running fine, until the energy drain starts to drop the voltage on the cap to a point too low for the rest of the power supply to function properly if the sag doesn’t end soon enough and the cap runs out of energy.
Depending on the load on the power supply at the instant that sag occurs, along with how low the magnitude of the sag is and its duration, are factors that determine whether the load “ride thru” or “trip off”. This is also true with the three-phase, full-wave rectifiers found in adjustable-speed drives. Not only is this “sensitive” equipment victim of power-quality phenomena, it is also a contributor to PQ problems with their harmonic currents. EC
BINGHAM, a contributing editor for power quality, can be reached at 732.287.3680.