Motor starts

At the exact moment that the motor starts, there is not a magnetic field in the winding. Thus, no resistance or impedance. Yes, the current can be 8 times the running current, but zero impedance means that you have zero volts (E=IR). Different power formulas exist but all have Volts and Amps as part of the equation. Thus I think that no matter which formula you look at, zero volts times inrush current equals zero power.

As time passe, a magnetic field builds, and this creates an impedance to current flow. These leads to a drop in current drops, and the voltage rises. As this drop in current and rise in voltage occurs, then power starts to appear. However, does the rate of change of current and voltage negate each other eliminate a spike. Or does this lead to a delayed power spike a few milliseconds after current starts to fall.

In my way of thinking, a power company is more interested in keeping power available to meet a facilities actual demand. Thus they look at the peak. However, they are not worried about the inrush since this is transitory and does not represent long term demand. Thus, they would have designed the demand meter to look at actual load, and not inrush.

When Starting a Motor across the line, you can draw up to 8 times the rating of the full load amps of the motor. A variable frequency drive or a soft starter limits the starting current due to a pre-charge circuit built into the drive or starter. VFD Drives are better because you can program the ramp time in the firmware.

Once I did a project at an Aluminum foil rolling mill to control the maximum demand. In India Maximum Demand for any given month is the highest kWh usage during any 1/Hr during the month. Exceeding this has a penalty of 20% on the entire month's billing. There are also 4 time zones during the day for each of which the MD is different.

In the project I did, the main rolling mill motors (from 800kW to 250kW) where the main contributors to such instantaneous power-demand surges. The PLC algorithm used a predictive estimation method of the likely maximum demand. In the event of the algorithm predicting the "likelihood" of exceeding the demand, selected heaters within Soak-Pits/Annealing furnaces (Typically having 18 heaters each of 100kW) would be switched off to pre-empt the transgression of Max Demand Limit. The strategy used for switching off was such that the goods under process were least affected.

Six priority levels were configured and Five different prediction methods to extrapolate demand were used. The rate of change of these extrapolated demands were given as input to PID Loops the output of which was used for determining the "likelihood" of exceeding the set MD limit . A consensus of at least 3 methods took a discrete decision to switch-off a "significant" load thus pre-empting the transgression.