Inrush and the laws of physics
Trying to explain power quality phenomena and even electrical concepts in general often results in attempts to translate them into a different domain, such as the mechanical world. Trying to come up with the mechanical equivalent of VARs (volt ampere reactive) is still elusive. However, the concept of inrush current is similar to the Newtonian laws of physics, which state that a body at rest tends to stay at rest; and a body in motion tends to stay in motion, unless acted upon by an external force. The parallel in the electrical world is that inductive devices tend not to like the current to change instantly; and capacitive devices tend not to like the voltage change instantly, unless provided with extra energy.
This concept applies to most electromagnetic devices, including motors and transformers, where the current is the electrical pushing force. To “get the ball rolling” requires a large amount of current in a relatively short time, called the inrush of current, into the motor or the transformer. Energizing a transformer means getting the magnetic fields re-established. This can result in higher-than-normal current levels for a short duration, especially if the transformer was last turned off with its core magnetized with one polarity, and the voltage is turned back on in the opposite phase or polarity. For motors, usually the bigger the motor, the larger the current; just like the bigger the rock, the bigger the force needed to make it roll. However, even small and medium horsepower motors can have inrush currents six to 10 times the normal steady state current levels.
Once again, Ohm’s and Kirchoff’s Laws come into play. Ohm’s Law states that Voltage = Current ¥ Impedance, and, in accordance with Kirchoff’s Law, the sum of voltages around a closed loop must be zero. If we assume 0.5 ohm source impedance and a 10A nominal on a 480V system, the inrush current that is six to 10 times the run or steady-state current can result in a drop of 30-50V, or a sag from nominal 475V at the load down to 430V. This is because the motor’s impedance looks somewhat like a short circuit when the rotor is stationary. Once the rotor starts turning, the current reduces and eventually goes to a much lower, steady-state value. However, if during the operation, a load change on the motor cause comes close to stalling or the locked rotor condition, then another sag can result for similar reasons.
These sudden changes in load current within facilities often result in voltage sags. These sags can result in other processes or equipment tripping offline, or creating product that is out of spec. Motors in one facility may be causing problems with equipment in another facility. In some cases, it may just cause the lights to blink. But with some newer HID lighting, this may result in complete darkness for several minutes until the lights restrike. In other situations, the sag resulting from motor starts can reduce voltage so low that the motors themselves may never properly start and come up to full speed. One motor starting up with a sag to 93 percent of nominal (7 percent reduction) may not seem like much and may not cause any problems in a facility. But suppose there are three 500 hp motors in the first floor of a plant. What will happen when two motors start up at the same time? Though not totally additive, the net result would be a much deeper sag to approximately 85 percent of nominal. These sags can also last many cycles, not just three to six cycles that are typical when faults occur on a distribution feeder. Whereas most equipment in the facility would have operated through the first sag, the second sag would most likely cause problems.
The severity of the effect of the sag depends on a number of variables, but measurements usual start with the voltage and current waveforms. In the example of a 10hp motor starting, the sag of the voltage and the inrush current is shown in the figure above, lasting about 2.5 seconds. The envelope of the motor start current has a exponentially decaying curve, as seen in the voltage and current rms time plot of our figure.
Since motors are said to consume the majority of the electricity in the United States, inrush problems can be found in many facilities. Checking the inrush data takes just a few seconds and can determine whether they can cause a power quality-based problem in the facility . (And if anyone has a suggestion of a mechanical model for the 90-degree phase shift and frequency dependent impedance in VARs, I would really like to hear it.) EC
BINGHAM, a contributing editor for power quality, can be reached at 732.287.3680.