HARMONIC FILTERS, APLCS, AND UPQCS

One means of ensuring that harmonic currents of nonlinear components will not unduly interact with the remaining part of the power system is to place filters near or close to nonlinear loads. The main function of a filter is either to bypass harmonic currents, block them from entering the power system, or compensate them by locally supplying harmonic currents.

Due to the lower impedance of the filter in comparison to the impedance of the system, harmonic currents will circulate between the load and the filter and do not affect the entire system; this is called series resonance. If other frequencies are to be controlled (e.g., that of arc furnaces), additional tuned filters are required.

Harmonic filters are broadly classified into passive, active, and hybrid structures. These filters can only compensate for harmonic currents and/or harmonic voltages at the installed bus and do not consider the power quality of other buses. New generations of active filters are active-power line conditioners that

are capable of minimizing the power quality of the entire system.

Passive filters are made of passive components (inductance, capacitance, and resistance) tuned to the harmonic frequencies that are to be attenuated. The values of inductors and capacitors are selected to provide low impedance paths at the selected frequencies. Passive filters are generally designed to remove one or two harmonics (e.g., the 5th and 7th).

They are relatively inexpensive compared with other means for eliminating harmonic distortion, but also suffer from some inherent limitations, including:

1. Interactions with the power system;

2. Forming parallel resonance circuits with system impedance (at fundamental and/or harmonic frequencies). This may result in a situation that is worse than the condition being corrected. It may also result in system or equipment failure;

3. Changing characteristics (e.g., their notch frequency) due to filter parameter variations;

4. Unsatisfactory performance under variations of nonlinear load parameters;

5. Compensating a limited number of harmonics;

6. Not considering the power quality of the entire system; and

7. Creating parallel resonance. This resonance frequency must not necessarily coincide with any significant system harmonic.

Passive filters are commonly tuned slightly lower than the attenuated harmonic to provide a margin of safety in case there are some changes in system parameters (due to temperature variations and/or failures). For this reason filters are added to the system starting with the lowest undesired harmonic.

For example, installing a seventh-harmonic filter usually requires that a fifth-harmonic filter also be installed. Designing passive filters is a relatively simple but tedious matter. For the proper tuning of passive filters, the following steps should be followed:

8. Model the power system (including nonlinear loads) to indicate the location of harmonic sources and the orders of the injected harmonics. A harmonic power (load) flow algorithm should be used; however, for most applications with a single dominating harmonic source, a simplified equivalent model and hand calculations are adequate;

9. Place the hypothetical harmonic filter(s) in the model and reexamine the system. Filter(s) should be properly tuned to dominant harmonic frequencies; and

10. If unacceptable results (e.g., parallel resonance within system) are obtained, change filter location(s) and modify parameter values until results are satisfactory.

In addition to power quality improvement, harmonic filters can be configured to provide power factor correction. For such cases, the filter is designed to carry resonance harmonic currents, as well as fundamental current.

Active filters rely on active power conditioning to compensate for undesirable harmonic currents. They actually replace the portion of the sine wave that is missing in the nonlinear load current by detecting the distorted current and using power electronic switching devices to inject harmonic currents with complimentary magnitudes, frequencies, and phase shifts into the power system.

Their main advantage over passive filters is their fine response to changing loads and harmonic variations. Active filters can be used in very difficult circumstances where passive filters cannot operate successfully because of parallel resonance within the system.

They can also take care of more than one harmonic at a time and improve or mitigate other power quality problems such as flicker. They are particularly useful for large, distorting nonlinear loads fed from relatively weak points of the power system where the system impedance is relatively large. Active filters are relatively expensive and not feasible for small facilities.

HIGH PASS FILTERS BASIC INFORMATION AND TUTORIALS

Swapping the cap and the resistor in the low-pass circuit creates another type of circuit called a high-pass filter. Using your now supreme powers of deduction and intuition, you are thinking to yourself, “I’ll bet that means the circuit passes high frequencies while blocking low ones.

 ” You are correct, and the circuit looks like the one in Figure 2.34 .Hopefully, after our discussion on the low-pass circuit, the operation of this one is clear.

The cap acts like a larger resistor at low frequencies, making the voltage divider knock down the output. At higher frequencies the cap passes more current as it becomes a short, causing a higher voltage at the output.

The inductor version of this circuit looks like Figure 2.35

As you might have suspected, this fi lter is the inverse, circuit-wise, of the RC high-pass filter. Another little bit of serendipity is the fact that the half-voltage output point 29 is also at 1/tau ( tau means time constant generically, whether referring to an RC or an RL circuit), just like the low-pass filters.

To sum up, the high-pass and low-pass fi lters take advantage of the frequency response of either a capacitor or an inductor. This is done by combining them with a resistor to create a voltage divider that attenuates the unwanted frequencies while allowing the desired ones to pass.

Some cool things happen when we put the two reactive elements together. You can create notch and band pass filters where a specifi c band of frequencies is knocked out, or a specific band is passed while all others are blocked.

The phenomenon of resonance also occurs in what is called a tank circuit, where you have a capacitor combined with an inductor. The tank circuit will oscillate current back and forth from one component to the other.

LOW PASS FILTERS BASIC INFORMATION AND TUTORIALS

WHAT ARE LOW PASS FILTERS?

Consider the circuit shown in Figure 2.32 . Note similarities to the RC circuit that we used to first understand the effects of a capacitor. The difference is that now we are going to apply an AC signal to the input rather than the step input we applied before.

This circuit is known as a low-pass fi lter, and all you really need to know to understand it is the voltage divider rule and how a capacitor reacts to frequency. If this were a simple voltage divider, you could figure out, based on the ratio of the resistors, how much voltage would appear at the output.

Remember that the cap is like a resistor that depends on frequency and try to extrapolate what will happen as frequency sweeps from zero to infinity. At low frequencies the cap doesn’t pass much current, so the signal isn’t affected much.

As frequency increases, the cap will pass more and more current, shorting the output of the resistor to ground and dividing the output voltage to smaller and smaller levels. There is a magic point at which the output is half the input.

It is when the frequency equals 1/RC. You might have noticed that this is the inverse of the time constant that we used earlier when we first looked at caps. Kinda cool when it all comes together, isn’t it? This is known as a low-pass filter because it passes low frequencies while reducing or attenuating high frequencies. You can make a low-pass filter with an inductor and resistor, too.

Given that the inductor behaves in a way that is opposite of a capacitor, can you imagine what that might look like? Have a look at Figure 2.33 .

That’s right; you swap the position of the components. That’s because the inductor (being the opposite of a cap) passes the lower frequencies and blocks the higher frequencies. It performs the same function as the low-pass RC circuit but in a slightly different manner. You still have a voltage-divider circuit, but instead of the resistor-to-ground changing, the input resistor is changing.

At low frequencies the inductor is a short, making the ground resistor of little effect. As frequencies increase, the inductor chokes 28 off the current, reacting in a way that makes the input element of the voltage divider seem like an increasingly large resistance.

This in turn makes the resistor to ground have a much bigger say in the ratio of the voltage-divider circuit. To summarize, in the low-pass fi lter circuits, as the frequencies sweep from low to high, the cap starts out as an open and moves to a short while the inductor starts out as a short and becomes an open.

By positioning these components in opposite locations in the voltage-divider circuit, you create the same filtering effect. The ratio of the voltage divider in both types of fi lters decreases the output voltage as frequencies increase.

All this lets the low frequencies pass and blocks the high frequencies. Now, what do you suspect might happen if we swap the position of the components in these circuits?