The categories or labels used to describe the plethora of power quality phenomena that can occur on electrical systems have not changed much in the past decade or two. Measurement methods and acceptable limits for most systems have changed. What is still referred to as “the GE curve” (a graphical way to look at the rate of variations in the root-mean-square (rms) voltage to try to determine the impact on incandescent lighting) has been replaced by a mathematically intense algorithm that produces a new parameter called perceptibility short term (Pst). If Pst is more than 1, complaints about flicker are likely; less than 1, not so likely. It sounds pretty simple and is commonly used in the rest of the world, but there is some resistance to it being widely adopted in the United States.


Unbalance is another category of power quality phenomena that has been around for years. Until recently, it was based on the NEMA MG-1 definition of maximum deviation of any phase value from the average of the three-phase values, whether it is voltage or current. Now, the more accurate and “universal” method (used in rest of world and only somewhat in United States), uses mathematical transforms that generate parameters called sequence components. Unbalance is defined as the sum of the negative sequence components divided by the sum of the positive and zero sequence divided by positive sequence. But, the need and applications remain the same.


If one steps back and looks at the energy spectrum of the phenomena as outlined in the IEEE 1159 Table 4.1, there has been a gap of undefined voltage variations. Small variations in rms voltage occurring in the 0–30-hertz (Hz) range are the domain of flicker. Reduction in rms voltage below 90 percent of the nominal voltage are sags, above 110 percent are swells, and below 10 percent are interruptions. These changes in voltage can last from a half-cycle to a minute. Very fast changes are the realm of transients, with a host of labels to describe their characteristics. But, what about rms voltage variations that occur between 3–10 percent of the nominal?


Once again, the international standards have a lead in defining this phenomena, called rapid voltage changes (RVC). Norway’s Norwegian Water Resources and Energy Directorate (NVE) has been a leader here, with the national regulatory agency defining this phenomena nearly 10 years ago. The latest draft of edition 3 of IEC 61000-4-30 has formally addressed RVC, though, as with other standards, the definition isn’t exactly the same. Given the legacy of the NVE definition, it’s the one that we will use here.


The figure above illustrates the concept behind an RVC. The voltage at time t0 is stable at a value of V0. Stable is defined as staying within ±0.5 percent of the steady-state value, called the “pre-event nominal.” The rms voltage is updated every half cycle, though it is computed over one complete cycle. In this example, the voltage then decreases at T1, though a sudden increase in voltage is also an RVC. Such a step change in voltage is often caused by a load starting up, such as a heat pump or heating, ventilating and air conditioning unit in a residential location. In most cases, the change is not enough to cause other equipment to trip offline or malfunction. 


That’s where the sag threshold is usually set, but it can indicate a pending problem. Some utilities refer to them as “blinks” since they cause the incandescent luminaires of the past to have their lumination drop for an instant.


The rms voltage drops to a minimum value V1 at time T2 and then recovers to a new steady-state value V2 at time T3. But since the voltage may fluctuate again and again, making it a flicker phenomena, the RVC definition requires that the voltage remain stable for a period of time. Again, the rms voltage remains stable, being ±0.5 percent from the V2 value. 


In the NVE, this stabilizing time window is 1 second. The maximum voltage deviation is V0–V1, while the deviation from the pre-event nominal or steady-state value to the post-event rms voltage is V0–V2, the “stationary deviation.” The maximum and stationary deviation values, along with the duration T3–T1, are the attributes or characteristic values that are associated with an RVC.


In general, the trigger limits for recording the change in voltage as an RVC is between 3–5 percent for the maximum deviation voltage. If the voltage crossed the limit for a sag or swell, the event wouldn’t be classified as an RVC. European regulatory agencies have set limits on how many RVCs of a specified magnitude are allowed on a daily or weekly time period. In the United States, this parameter is still relatively new, and, like other power quality phenomena (except for perhaps harmonics and interruptions), having legal or regulatory limits for such utilities or consumers that can be penalized is a rarity so far. Time will tell if this new category will gain more prominence as its inclusion in power quality monitors becomes more common, and users become more familiar with its value in fully characterizing an electrical system’s performance.