Block Demand vs Sliding-Window Demand — How Your Maximum Demand Is Measured

Block demand — fixed intervals, successive windows

Block demand divides the billing period into successive, non-overlapping intervals of a set length — typically 15 minutes, though some DISCOMs use 30-minute intervals. The demand for each interval is the average power consumed during that specific window: total energy consumed in the interval divided by the interval duration in hours.

The maximum demand for the billing period is simply the highest of all the interval averages. A 720-hour month with 15-minute intervals produces 2,880 readings; the single highest of those 2,880 interval averages is the maximum demand on which your demand charges are billed.

The key characteristic of block demand is that the intervals are fixed and successive: the clock resets every 15 minutes at the same wall-clock time. Interval one might be 07:00–07:15, interval two 07:15–07:30, and so on. Each event that occurs during a given block contributes to exactly one interval average.

A consequence of this structure is that a short, sharp spike that straddles a block boundary — beginning at minute 14 of one block and ending at minute 3 of the next — has its energy split across two intervals. Each interval sees only part of the spike, so the averaged demand for each is lower than if the entire spike had fallen within a single interval. Under block metering, the timing of a spike relative to the block boundary affects what maximum demand it produces.

Why the method matters — same load, potentially different bills

For load profiles that are smooth and evenly spread across time, block and sliding-window demand produce nearly identical results. For facilities with sharp, short-duration spikes — large motor starts, batch process events, or compressor inrush cycles — the two methods can diverge meaningfully.

The magnitude of the difference depends on two factors: how sharp the spike is (high peak power for a short duration is harder to “split” beneficially across block boundaries) and how close the spike’s timing is to a block boundary. In the worst case — a spike that is perfectly centred on a block boundary — the block method sees roughly half the spike energy in each of two intervals, while the sliding-window method sees the entire spike in one window. The difference in recorded maximum demand could be substantial.

In practice, most facilities do not have spikes that are consistently timed to exploit block boundaries. But the difference is non-trivial when it occurs, and it is invisible to energy managers whose monitoring system uses the opposite calculation method from the utility meter. The facility’s monitoring shows a lower peak than the utility meter recorded, the alert did not fire, and the exceedance penalty appears on the bill with no prior warning.

Matching your monitoring meter to your DISCOM — why it is not optional

A monitoring meter set to block demand will produce readings that align with your utility bill only if your utility meter is also configured for block demand. If there is a mismatch, you have two meters tracking the same load and producing systematically different maximum demand readings — and you will trust the wrong one.

This matters most for pre-crossing alerts, which are the real-time safety net for demand management. An alert threshold set at, say, 85% of contract demand is meaningful only if the monitoring meter is measuring demand the same way the utility meter is measuring it. If the monitoring meter under-reads demand relative to the utility meter, the alert fires at the wrong moment — too late, or not at all before the utility meter records the exceedance.

The Titan energy meter (Class 0.5S per IEC 62053-22) supports both block and sliding-window demand calculation. The demand method is set during commissioning to match the DISCOM’s utility meter configuration. With the correct setting, Titan’s live demand reading tracks what the utility meter is accumulating in real time — so when Titan signals 90% of the contract demand limit, the utility meter genuinely reads 90%, and the operator has an accurate picture of remaining headroom.

For multi-site deployments where different DISCOMs use different methods, maximum demand monitoring with per-site configuration ensures that each site’s monitoring matches its local utility meter — rather than applying a single configuration globally and accepting the mismatch at sites where the method differs.