Summer Peak Demand Charges: A Warehouse Worked Example

In short: most commercial demand tariffs are set by the single worst 15 or 30-minute interval of the billing period, and a run of 40-degree days is exactly when that worst interval happens. Summer peak demand charges from one hot week can land on the bill for a month, and on a ratcheted tariff for the following twelve. On the worked example below, a heatwave lifts a warehouse’s peak from 180 kW to 260 kW; at $16/kW/month that single week is worth $1,280 extra in the month it happens and $15,360 extra over the year if the tariff ratchets. Solar helps at 2 pm and does very little by 6 pm, which is exactly when a heatwave peak tends to land.

Send this to a customer before summer, not after. Once the ratchet has locked in a bad month, there is nothing left to model; the number is already on twelve invoices. The window to do something about a site’s peak demand season exposure is the shoulder months, while last year’s interval data is still sitting there answering the question.

We covered demand charges generally in Demand Charges and Solar for Australian Business. This piece is narrower: what specifically happens to a demand charge during a heatwave electricity demand event, what a battery on demand duty does to that spike with the arithmetic shown, and how to read a customer’s own interval data to find their exposure before the first 40-degree week of the season arrives.

How One Hot Week Sets the Bill

A demand charge is not an average of the billing period. It is set by the single highest interval the meter records, usually 15 or 30 minutes long. On a normal week, HVAC cycles, occupancy, and equipment schedules keep that peak fairly predictable. During a heatwave, every one of those loads runs harder at the same time: compressors work against a bigger temperature difference, chillers stay on longer, and refrigeration and cool-room equipment lose their overnight recovery window because the building never fully cools down. The peak that results is not a small step up from a normal day; it is often the highest reading of the entire year, set inside a single week.

Two separate things then determine how much that week costs. First, whether the tariff is billed month by month, in which case the elevated peak only affects the invoice for the month it happened in. Second, whether the tariff carries a demand ratchet, where the charge for the next twelve months is set by the highest peak recorded in that trailing window, not by the current month’s own peak. Under a ratchet, one 40-degree week in January can still be setting the bill in November. Whether a site’s tariff ratchets, and over what window, is on the bill or the network tariff schedule; it is worth checking before assuming either way.

Worked Example: A 2,000 m² Warehouse Through a Heatwave Week

Assume a 2,000 m² warehouse with rooftop HVAC and a normal-week peak of 180 kW, on a demand tariff of $16/kW/month. In a heatwave week the site’s peak climbs to 260 kW: the same equipment, working harder against the heat, for longer hours.

Normal month: 180 kW × $16/kW/month = $2,880
Heatwave month: 260 kW × $16/kW/month = $4,160
Extra cost that month: $4,160 − $2,880 = $1,280

That is the entire pitch for looking at this before summer, not after. The same physical event (one hot week) is worth $1,280 on a monthly tariff and $15,360 on a ratcheted one. A customer who has never had this explained to them will assume the number resets in February. On a ratchet, it does not.

Why Solar Helps at 2 pm and Does Little at 6 pm

Solar output tracks the sun, not the temperature. It peaks around the middle of the day and tapers through the afternoon to zero at sunset. A warehouse’s heat load does not follow the same curve. Ambient temperature keeps rising into the afternoon, the building envelope and roof have been absorbing heat for hours, and HVAC systems are often working hardest in the late afternoon and early evening, well after solar output has started its decline. It is a variant of the same mismatch covered for network peak periods in our PDRS and ESS article, where the demand window the scheme pays for runs from the afternoon into the evening for the same reason.

This is why a solar-only system can genuinely help a warehouse’s midday load and still do close to nothing for the interval that actually sets the demand charge. If the heatwave peak in the worked example above lands at 2 pm, a well-sized array can shave real kilowatts off it. If it lands at 6 pm, once the sun is low and the building has spent all afternoon heating up, solar production is a fraction of nameplate and the array is doing almost nothing for that specific interval, even though it is still saving energy charges across the rest of the day. Whether solar helps this particular charge is a question about the timing of the peak, not the size of the array.

What a 60 kW / 120 kWh Battery Does to That Spike

A battery on demand duty does not need to cover the whole load; it only needs to shave the difference between the heatwave peak and a target the site is comfortable with, for as long as that peak interval lasts. Take a 60 kW / 120 kWh battery reserved for demand shaving on the warehouse above:

Duration at full output: 120 kWh ÷ 60 kW = 2 hours of discharge, enough to cover a typical afternoon HVAC peak.
Heatwave peak with battery: 260 kW − 60 kW = 200 kW
Heatwave month cost with battery: 200 kW × $16/kW/month = $3,200 (down from $4,160, a saving of $960 that month)
If the tariff ratchets, the billed peak for the full year drops from 260 kW to 200 kW: (260 kW − 200 kW) × $16/kW/month × 12 months = $11,520 saved over the year

Two things to flag when quoting this to a customer. First, the battery has not made the warehouse’s summer peak disappear; it has pulled it down to 200 kW, which is still above the 180 kW normal-week peak. A battery reduces exposure; it does not guarantee the ratchet never fires again unless it is sized to cover the full spike every time. Second, the 2-hour duration is the binding constraint, not the 60 kW rating: if a site typically holds its worst interval for three or four hours through a heatwave afternoon, either the kWh needs to grow or the discharge needs to be reserved specifically for the worst part of that window rather than run flat across all of it. This is the same demand-duty sizing logic covered in more depth in our general demand charges article; the seasonal point here is that the worst-case discharge window to size against is the heatwave peak, not a typical day.

In Amperage, this is the same hourly dispatch simulation that runs any battery scenario against the site’s own load; there is no separate “summer mode.” Feed in a full year of interval data that includes at least one heatwave week and the simulation will size the battery’s demand-duty reserve against the load shape that actually shows up, rather than against an assumed average day.

Operational Measures That Cost Nothing

Before any capital is spent, a handful of scheduling changes can pull real kilowatts off a heatwave peak for free:

• Pre-cooling: run HVAC harder in the morning, while it is still cheap and before the peak window, to cool the building mass ahead of the afternoon heat load. This borrows against the coming peak instead of fighting it in real time.
• Staggered starts: if multiple compressors, chillers, or cool rooms cycle back on around the same time (common after a lunch break or a scheduled defrost), a few minutes of offset between each unit’s restart avoids stacking their inrush and running load into the same 15-minute interval.
• Setpoint discipline during the worst hours: a 1–2°C setpoint relaxation for the two or three hours around the historical peak window costs comfort, not capital, and is easy to wind back once the heatwave passes.
• Deferring non-essential load: forklift charging, non-urgent batch processes, and discretionary equipment can be scheduled outside the known peak window entirely, the same load-shifting logic covered in the general demand charges article above.

None of this needs a quote. It needs someone to look at last summer’s data and identify which hour actually mattered.

Reading Last Summer’s Interval Data to Find the Exposure

Every business with a smart meter has this answer sitting in its own interval data already. The exercise is short:

1. Pull twelve months of 15 or 30-minute interval data from the retailer or distributor, if it has not already been captured during onboarding.
2. Sort by kW, not by date, and pull out the top 10–20 highest intervals across the year.
3. Check the date and time of each one. If most cluster in December to February and land in the early-to-mid afternoon, solar is doing real work on those intervals. If they cluster in the same months but land from late afternoon into early evening, solar is not the fix and the battery or operational measures above are where the value is.
4. Compare the single highest interval of the year against a typical summer day’s peak. The gap between the two is roughly what a heatwave is costing on top of an ordinary hot day, and, on a ratcheted tariff, what is riding on the bill for the following twelve months.

This is exactly the load-profile check described in Demand Charges and Solar for Australian Business and in the consumption upload docs, applied specifically to the hottest week of the year rather than the average one. Uploading that data into an Analysis project surfaces the same peaks automatically, without a manual sort.

Summer Readiness Checklist

Send this to a customer, or run through it yourself before quoting, ahead of the season:

• Confirm whether the site’s demand tariff ratchets, and over what trailing window.
• Pull last summer’s interval data and identify the time of day the top peaks actually occurred.
• Note the demand rate ($/kW/month) currently on the bill; it is the multiplier for every number above.
• Check whether the peak-hour window overlaps with solar generation hours or sits after it.
• Review HVAC and equipment start-up schedules for stacking risk during the hottest hours.
• If a battery is on the table, size the discharge duration against the heatwave peak’s length, not a normal day’s.

Where to Go From Here

A heatwave week is not an anomaly to explain away after the invoice arrives; it is the predictable event that a commercial demand tariff is designed to bill hardest for. The sites worth flagging before this peak demand season are the ones whose interval data shows peaks sitting in the late afternoon, outside solar hours, on a ratcheted tariff. Those are the customers a battery-on-demand-duty conversation is worth having with now, while there is still time to act on it.

Demand Charges and Solar for Australian Business →

How TOU tariffs interact with solar →

How Amperage’s hourly dispatch modelling works →