The frequency stability of the interconnected electric grid in North America is a marvel of engineering. As shown in Figure 1, the deviation is rarely beyond 0.05 Hz from the nominal 60 Hz. In fact, during a given hour time period, this deviation is more like +/-0.02 Hz from the nominal value. Over several months, we were able to track the frequency at three monitoring sites separated by more than a thousand miles to be basically identical at any point in time. This kind of synchronization and coordination across miles and miles of wire may tend to lull us into thinking that frequency is not on the list of potential power quality phenomena to be concerned about.
Except during fault conditions, this is usually the case when powered from the grid. However, when running off of an alternative power source, there are two changes in the characteristics of the power source that we should consider. The first is the stability of the frequency during load changes. Though the mechanical-to-electromagnetic transformation of the generation process is similar on a 10kVA diesel-driven generator and a 300MVA hydro- electric power plant, there is one significant difference—mechanical inertia, which translates into frequency inertia.
A step increase in current draw from the loads will slow the generator down momentarily, and a sudden decrease in load will tend to speed the generator up for a short duration. The result is often a much wider frequency deviation in load changes than when running of the electric grid. For example, a typical 60kVA diesel generator from a well-respected manufacturer has a published frequency regulation spec of 3-5 percent for no load to full load, and a 0.33 percent regulation under steady state conditions. This is an order of magnitude different from the aforementioned data collected from the electric grid.
Other backup types of equipment, such as a UPS, may also have a change in the frequency stability. Figure 2 shows such a deviation change on the output of a UPS when operating on standby power. There are some synchronous processes that will operate differently under such frequency variations. The most obvious are timing circuits that use the zero crossings of the voltage sine wave to be 16.6666 msec per cycle for a 60Hz power frequency. On old disk drives that had synchronous motors turning the disks, writing at one speed and then reading back at another speed may not be viable.
Another aspect that deserves consideration is the voltage stability during load changes. The source impedance of the typical wall outlet is less than 1 ohm. A 0.5 ohm impedance on a 120V circuit will experience sag to only 95 percent of nominal when the current increase from 1A to 10A. This generally will not affect most equipment powered from the circuit. A backup generator’s source impedance is likely to be significantly higher. If it were just 2 ohms, the same step change in current would reduce the remaining voltage down to 100V. In addition, the sudden change is load may also affect the voltage output of the generator itself. With the generator mentioned in the frequency stability example, this voltage regulation under no to full load changes was rated quite good, +/-1 percent. However, the combination of voltage and frequency reduction with a load increase (and vice-versa) with a “less-stiff” generator may wreak havoc with loads that appear to function just fine when powered from the grid.
One last concern is the transfer mechanism. When the system protection equipment operates during fault conditions on the electric grid, there may be a small phase shift. When switching from a generator back to the grid, there should be a mechanism that verifies that the frequency and the phase angle are synchronized before switching back if there are any loads that would “object” to a sudden phase shift (or worst case, phase reversal). Single phase, rectified input, switch-mode power supplies that turn AC into DC would operate through such just fine. However, an electric induction motor wouldn’t be too happy with a 180-degree phase shift.
Not all facilities with backup generators will experience any of the aforementioned problems. Many well designed systems operate quite well and provide the necessary quality of supply that the loads in the facility need. But there are plenty of examples, when the “hertz can hurt” if not properly maintained from the power source. EC
BINGHAM, a contributing editor for power quality, can be reached at 732.287.3680.