Whether it came from Heraclitus in 470 B.C. or François de la Rochefoucauld in the 17th century, the adage “the only thing constant is change” usually directly affects the power quality level in a facility. Unless there is a conscious effort when changes are made to equipment and/or infrastructure, it often isn’t for the better. However, a few prechange tasks with some postchange follow-up work can produce cash for all involved.

Before changing anything in a facility, it’s important to inspect its current status, which involves getting one-line diagrams updated accurately along with a list of equipment connected to each circuit where there will be modifications made. Though the ideal would be to conduct measurements on each piece of equipment, it is more likely that time will permit only baseline monitoring of the voltage, current and basic power quality parameters (harmonics, rms variations, transients) on the affected circuits. Don’t overlook recording the simple measurements of line-to-line, line-to-neutral (L-N) and neutral-to-ground (N-G) voltages, as applicable. Also remember that the current total-harmonic-distortion is somewhat meaningless. However, the rms harmonic current levels, as compared to the circuit’s capacity, is important.

If the existing readings already indicate potential problems or borderline conditions—such as N-G voltage readings over 1 volt (V) or total harmonic voltage distortion (Vthd) over 5 percent—the source of these problems should be investigated and fixed before making any modifications. Say, for example, a small manufacturing facility was going to add some additional server equipment in the computer room. The IT manager told you that he had to be careful which outlet the equipment was to be plugged into, as previous equipment had malfunctioned in certain outlets. In addition, the circuit breaker for that room had tripped recently without any apparent cause.

A quick measurement of the L-N and N-G voltages in the facility found that the N-G voltage was greater than 2V at all of the outlets in the computer room but was under 1V elsewhere. A power quality monitor measuring just the L-N and N-G voltages found that sharp decreases in the L-N voltage corresponded to increases in the N-G voltage. Some of the decreases were severe enough to be classified as sags, although no malfunctions of the information technology (IT)equipment were noted during that time as most of the equipment operated off of uninterruptible power supply units.

For single-phase receptacles, an indication that the sag in L-N voltage is caused by a load, and not the utility or electrical source, is when there is an increase or “swell” in the N-G voltage corresponding to the decrease in L-N voltage, typically a sag 50 percent more intense than the swell. For example, a 10V reduction of L-N would have a 5V increase in the N-G voltage, which was the case here. Further investigation revealed that the 20-ampere (A) circuit was loaded near capacity, and there was a receptacle outside of the computer room on the same circuit where a piece of manufacturing equipment with a heating element was used occasionally. If they had added any equipment before fixing these problems, the change wouldn’t have been positive, and most likely, the new IT equipment would have been blamed for the pre-existing problems.

You should also look at the harmonic levels in the prechange measurements. If the Vthd levels are above 5 percent, check the individual harmonic values for the dominant harmonics. If the 3rd is the largest, followed by the 5th, then 7th and so on with the odd harmonics, the predominant loads are likely single-phase with rectified input switch mode power supplies, as found in most IT equipment. If the dominant harmonic is the 5th and 7th, then 11th and 13th, and so on, look for three-phase loads, such as adjustable speed drives that also have nonlinear power supplies. If the levels are high on one circuit and not another, determine if it is feasible to relocate some of the “polluting” loads to distribute the harmonic currents to levels that would balance out the harmonics.

Balance or lack of balance (unbalance) is another parameter that is easy to check. Voltage unbalance usually is the result of current unbalance, though an incorrect tap setting on the transformer feeding the facility also is a possibility. As with the harmonic levels, redistribute the load where possible to achieve a voltage unbalance of less than 1 percent.

Next, check the nameplate rating of the loads to be added to verify that the circuits have adequate capacity for them. The expected harmonic contribution from any nonlinear loads should also be reviewed for their possible impact. Any large polluters should be on their own circuit and not on the same circuit as any mission-critical equipment, such as the servers.

Once the new circuits and equipment have been added and are operational, repeat the prechange tests to verify that the new quality of supply has not been compromised. In fact, if the aforementioned mitigation techniques were applied before the change, the quality of supply should be improved, and the chance is reduced of power quality phenomena causing any process interruptions with negative financial impacts.

The facility gets the change that it needs and gets to add more dollars to the bottom line. And, it is another opportunity for electrical contractors to add to theirs as well.


BINGHAM, a contributing editor for power quality, can be reached at 732.287.3680.