Power Quality Complaints: Is It the Utility or the Load?

I have investigated two power quality complaints that turned on whether the problem came from the utility or from the load. In the first, an agricultural operation with robotic and power-electronic equipment reported that "dirty power" from the utility was damaging equipment and affecting its animals. In the second, a rural customer reported harmonics on a distribution feeder that also served two large industrial plants. The regulator's service rules required the utility to show that its delivered power met the applicable limits, and the answer depended on measuring at the right place with the right method.

Simplified Explanation

A utility is responsible for the voltage it delivers at the point where its system meets the customer's, the point of common coupling (PCC). A customer is responsible for the current its equipment draws. Most power quality problems are created by loads, either the customer's own or a neighbor's on the same feeder. A properly designed measurement separates the two.

Measure it so the answer holds up

1. Measure at the PCC. That is where the utility's obligations and the IEEE Std. 519 limits apply. For a customer served from a dedicated transformer, the PCC may be on the primary side of that transformer. Identify the PCC before the meter goes in.
2. Use a Class A meter. Use an IEC 61000-4-30 Class A instrument so that rms, harmonic, unbalance, and flicker values are computed by a standardized method that another expert can reproduce.
3. Record for at least a week. Both IEEE Std. 519-2014 and the flicker standard evaluate compliance with percentiles over a weekly window.
4. Take a baseline. In one study a resistive load bank was connected with the customer's load offline, to show what the utility supply looked like without the customer's equipment.
5. Measure in more than one season. In the feeder study, summer measurements captured the industrial plants at heavy load, and winter measurements captured the utility's worst-case baseline (light industrial load, heavy residential load).
6. Log the load state. Record when large loads start, stop, and change mode so that events in the data can be matched to equipment.

Read the results against the right limit

Compare voltage regulation against the service rule and the ANSI C84.1 ranges for that class of service. ANSI C84.1 recommends that the utility limit voltage unbalance to 3%, measured at the meter under no-load conditions, so brief excursions with load are not by themselves a violation.

IEEE Std. 519-2014 (Table 1) sets the voltage distortion limits by bus voltage at the PCC:

The limits are evaluated statistically. The weekly 95th percentile of the 10-minute values should be below Table 1, and the daily 99th percentile of the 3-second values should be below 1.5 times Table 1. A 100th-percentile excursion above the limit therefore does not mean non-compliance when the governing percentiles are below it. Check which edition the state's service rules adopt. A rule that incorporates IEEE Std. 519-1992 applies 3% individual and 5% THD to every bus at 69 kV and below, including a 480 V service, and that edition allows the limits to be exceeded by 50% during start-ups or unusual conditions lasting less than an hour.

Flicker is evaluated the same way. IEEE Std. 1453-2015 recommends planning levels of Pst 0.9 and Plt 0.7 at medium voltage (0.8 and 0.6 at high and extra-high voltage), and compatibility levels of Pst 1.0 and Plt 0.8 at low voltage. The weekly 95% value should not exceed the planning level, and the 99% value may exceed it by a factor of 1 to 1.5 depending on system conditions.

Find where it comes from

• Harmonic signature. Elevated 5th and 7th harmonics are typical of six-pulse power electronic converters, such as variable frequency drives. In the feeder study the 5th harmonic exceeded its individual limit in summer even with one of the two industrial plants offline, which suggested the other plant as a likely source.
• Symmetrical components. In the agricultural study, decomposing motor inrush events into positive-, negative-, and zero-sequence components showed a very large negative-sequence current from on-site load. Negative-sequence current heats motor rotors and may explain motors failing or tripping on their own protection.
• Event direction. The captured events were inrush currents from on-site equipment, and the voltage dips during those events tracked on-site loads starting. None of the captured events originated outside the facility.

In both studies the utility met the public service commission's power quality requirements at the measured location. Voltage THD was within the IEEE Std. 519 limit, and where an individual harmonic exceeded its limit, the measurements traced it to customer load. The other issues measured also came from loads: high inrush, unbalanced load currents, and harmonic-rich current draw.

What it means for customers and utilities

These mitigation options may or may not be appropriate for a particular site. The list is not exhaustive and is not in any particular order.

1. Identify the dominant harmonic source by measuring at individual pieces of equipment, then correct it at the source.
2. Apply harmonic filters tuned to the offending harmonic at the equipment that produces it, where practical.
3. Replace diode-bridge front ends with active (PWM) rectifiers where the equipment allows it.
4. Rebalance single-phase loads across the three phases.
5. Recheck neutral and grounding connections, including resistance and lug torque.
6. Protect sensitive equipment with a UPS or ride-through device.
7. On the utility side, where customer loads are not corrected, options include lowering source impedance (reconductoring or a larger or lower-impedance substation transformer) and utility-owned filtering. These require a system study before any decision.

What is commonly misread

A single handheld reading taken inside the facility does not show what the utility delivered at the PCC. When access to the customer's equipment or one-line diagram is refused, the report should say so and limit its conclusions to what was measured.

Standards referenced

• IEEE Std. 519-2014, harmonic control in electric power systems
• IEEE Std. 1453-2015, measurement and limits of voltage fluctuations and flicker
• IEEE Std. 1159, monitoring electric power quality
• ANSI C84.1, electric power systems and equipment voltage ratings
• IEC 61000-4-30, power quality measurement methods (Class A)
• The applicable state public service commission's electric service rules

Details have been generalized to protect client confidentiality.

Related Power quality investigation · Engineering services · All insights