Two plants on the same tariff, drawing the same number of units in a month, can get bills that differ by lakhs. Nothing is wrong with the meter. The difference is maximum demand, and it is the one line on an HT bill that most plants never actively manage.

Maximum demand is an average, not a spike

The common assumption is that maximum demand records the highest instantaneous current the plant ever draws. It does not. The meter integrates demand over a fixed block, 15 minutes on most Gujarat tariffs and 30 in some others, and records the average for each block. At the end of the billing cycle it keeps the single highest block and bills you on that.

That distinction tells you what to attack. A motor pulling six times full load current for four seconds barely moves a 15 minute average. Four motors started inside the same ten minutes, every morning, move it a lot.

It also explains why demand is billed in kVA rather than kW. kVA is what the utility’s transformers, cables and switchgear have to be sized for, whether or not your plant turns it into useful work. We covered the other half of that equation in the power factor penalty post.

The peak is usually a habit, not a load

When we look at interval data from a plant’s energy meter, the monthly peak almost always lands in the same place: shift change. Compressors restart, the chiller comes back on, the furnace gets charged, the welding bay opens up, and the canteen load arrives, all inside the same half hour, because that is when people arrive.

Nobody designed that peak. It is the sum of everyone’s routine. Which is why the cheapest fix is almost never equipment. It is sequencing.

What actually moves the number

In rough order of return on effort:

  • Stagger the starts. Spreading large motor and heater starts across 45 minutes instead of 10 costs nothing and often takes 10 to 15 percent off the recorded peak.
  • Correct power factor properly. On kVA billing, poor power factor inflates demand directly. An APFC panel that switches with the load, rather than a fixed capacitor bank sized for full load, is what holds the figure through the day.
  • Soft starters or VFDs on the worst offenders. These cut the starting current itself. Averaged over a 15 minute block the effect is smaller than people expect, but on plants where motors cycle frequently it compounds.
  • Fit a demand controller. It watches the running average inside the current block and sheds pre-agreed non-critical load before the block closes. Water heating, some lighting circuits, non-urgent pumping.
  • Review contract demand. Most plants have this wrong in one direction or the other. Exceed it and you pay penalty rates on the excess. Set it too high and you pay for capacity you never touch. Twelve months of billed demand tells you where it should sit.

Where the panel comes into it

Almost everything above needs two things from the switchgear: metering you can actually read, and control over what starts when.

A plant with one meter at the incomer knows its total and nothing else. Feeder level metering is what turns “our bill went up” into “the second compressor is starting at 8:40, with the chiller.” We build metering panels with that in mind, because the measurement is what makes everything after it possible.

Worth checking harmonics at the same time, since distortion inflates kVA without doing any work. Our post on harmonics goes into it.

Frequently asked questions

Is maximum demand billed in kW or kVA?

Most HT and LT industrial tariffs in Gujarat bill demand in kVA. kVA includes the reactive component, so a plant with poor power factor pays a higher demand charge for the same useful work. Your tariff schedule states which applies to your connection.

Does one short spike raise my bill for the whole month?

Only if it lasts long enough to lift the average of a full demand block. A four second inrush will not. A load that stays up for ten minutes inside a fifteen minute block will.

What is the difference between contract demand and maximum demand?

Contract demand is the capacity you have agreed to pay for. Maximum demand is the highest block the meter actually recorded in that billing cycle. You are billed on the recorded figure, at penalty rates once it crosses the contracted one.

Will a soft starter reduce my maximum demand?

It reduces starting current, which helps most where motors start and stop often. Where the peak comes from several loads running together rather than from starting, sequencing and power factor correction will do more for less money.

How do I find out where my peak is coming from?

Download the interval data from your energy meter, or ask the utility for the load survey, then match the highest blocks against the shift and process schedule. Feeder level metering inside the panel makes this much quicker, because it tells you which circuit and not just what time.