Welcome to EC&M's monthly forum where power quality experts Mark McGranaghan, vice president of Electrotek Concepts, and Mike Lowenstein, president of Harmonics Ltd., address your PQ problems and concerns.

Q. How can I find out how many voltage disturbances to expect from my supply system? Voltage sags and momentary interruptions cause expensive downtime, and I want to evaluate how to improve the situation. Are standards available that summarize the approximate number of these events?


McGranaghan's answer. Most electric utilities understand that voltage sags can affect many industrial processes. However, many don't track system voltage sag performance like they track reliability (utilities define reliability based on outages that last longer than five minutes). EPRI and Electrotek characterized average voltage sag performance for distribution systems in the United States as part of a power quality benchmarking project in the mid-'90s. Fig. 1 above summarizes the average number of voltage sags an industrial facility can expect to see over a one-year period. Of course, there are wide variations in the expected voltage sag performance from one utility to another and from one system location to another within a utility. The most common index for describing voltage sag performance is the System Average RMS Variation Frequency Index (SARFI). It must be qualified by the voltage sag severity of concern. For example, SARFI-70 is the number of voltage sags that occur per year where the minimum voltage drops to less than 70%. The national average is about 18 events per year. Compare that to the national average for the System Average Interruption Frequency Index (SAIFI) — the number of outages lasting longer than 5 min. The national average for SAIFI is about 1.3 outages per year. Ask your utility if they keep the SARFI statistics for the substation that supplies your facility. Note: The estimate should be based on at least five years of historical data because the performance can vary significantly from year to year.

Q. I noticed listings for 6- and 12-pulse drives when looking through a catalog for adjustable speed drives. What's the difference, and how do I know which one to choose?

Lowenstein's answer: One feature of all adjustable speed drives (ASDs) and nonlinear loads is that they change the AC voltage coming from the power source lines to DC voltage. ASDs do this by using a circuit called a bridge rectifier, which reduces the flow of harmonic currents.

In ASDs, the rectifier draws current in spikes or “pulses.” The number of current pulses per electrical cycle depends on the configuration of this rectifier and the associated components. Three-phase nonlinear loads use rectifier designs with multiples of six pulses (6, 12, 18, 24, etc). The higher the number of pulses, the more complex and expensive the rectifier. The choice of a rectifier design determines the harmonic spectrum of the current drawn by the load.

The harmonic current spectrum of a particular load doesn't depend on the end-use of the DC voltage. ASDs, electroplating rectifiers, UPSs, and battery chargers draw the same characteristic harmonic currents if they use the same rectifier design. You can determine the characteristic harmonic current spectrum of a rectifier using the following equation:

h = n x p +/- 1

where h is the harmonics number, n is the sequential integer, and p is the pulse number of the rectifier. To see how to use the equation, let's examine the 6-pulse rectifier. The first harmonic pair is 1 x 6 +/- 1 = 5 and 7. The next pair is 2 x 6 +/- 1 = 11 and 13. For a 12-pulse rectifier, the first harmonic pair is 1 x 12 +/- 1 = 11 and 13. The next pair is 2 x 12 +/- 1 = 23 and 25.

As you can see, as the pulse number of the rectifier increases, the lower order harmonic currents disappear. So selecting a high-pulse number rectifier is one method of eliminating harmonic currents without the expense of filtering. Instead of removing harmonic currents after they've formed, a high-pulse rectifier prevents the formation of some of these currents. At least one manufacturer of high-horsepower, medium-voltage drives uses 24- or 36-pulse rectifiers to eliminate lower-order harmonic currents.

Along with this benefit, however, comes a higher cost. The higher pulse rectifier systems are more expensive than a simple 6-pulse system. For these high-horsepower drives, the extra cost for the 24- or 36-pulse rectifier system is less than the cost of a 6-pulse rectifier plus the necessary filters. On the other hand, for a 50-hp drive, such a rectifier would increase the drive cost by several times and wouldn't make sense.

The type of drive rectifier system you specify depends on parameters like transformer size, drive size, single or multiple drives, system capacity, and harmonic standards or regulations.

For a single low-horsepower drive, the investment in a 12-pulse rectifier is seldom worthwhile. For multiple low-horsepower drives or a single higher horsepower drive, specifying 12-pulse rectifiers could be cost-effective. Whatever your situation, you can make an educated decision as to which type of rectifier system to specify for the drives you order by keeping in mind what the pulse number means in term of harmonic currents, what the relative costs are of 6- and 12-pulse drives, and the cost of harmonic filtering.

Q. I'm the electrical engineer for a plant that produces plastic bags. Sometimes we get disturbances that cause our lights to blink. These disturbances often also disrupt one or more of our 18 bag lines, which causes a major problem. Occasionally, all of the lines are interrupted, but in most cases, only a few lines are affected. Why do power disturbances only affect some of my processes?

McGranaghan's answer. Plastics extrusion lines can be very sensitive to voltage sags caused by faults located somewhere on the utility supply system. The fault could be on the utility's distribution circuit supplying the plant, on another distribution circuit supplied from the same substation bus, or somewhere on the transmission system. Many of these faults are caused by single line-to-ground faults (Fig. 2). Such voltage sags don't affect all three phases. In fact, the Distribution Power Quality (DPQ) benchmarking project performed by Electrotek for EPRI in 1994-1996 showed that 66% of voltage sags on distribution systems affect only one phase. This can influence whether or not a specific extruder line is affected. If the controls for that line are supplied from the affected phase, the controls might cause tripping of the line. (Sometimes protecting the controls can significantly improve the ride-through of plastics extrusion lines.) Other lines might not be affected. In other cases, the extruders themselves could feel the effect. Whether or not the extruders are affected can depend on specific conditions on the extruder like the firing angle or load, the phase shift during the sag, the instant that the sag is initiated, the voltage magnitude, the duration, and the unbalance during the sag.

The response of each extruder can be different depending on the particular operating conditions at the time of the sag as evidenced by the fact that some of your lines are affected and others aren't for the same voltage sag. There are technologies available for protecting your extruders so they can ride through these events, but they all come at a cost. You have to evaluate the costs of these solutions against the costs that you are incurring every time the plastics lines are disrupted. In order to perform the economic analysis, you'll need an estimate of the expected number of voltage sags at your facility.

If you find yourself perplexed by a power quality problem, just e-mail your question to Editorial Director John DeDad at jdedad@primediabusiness.com. He'll get your question in front of our power quality experts, who'll answer as many questions as they can each month in this column.