# Pre-Cooling and HVAC Scheduling Against the Peak > Buildings store heat. Running the cooling plant harder before the expensive window and coasting through it turns that thermal mass into free storage. Section: Peak Demand Management Author: Nil Masferrer Jiménez Published: 2026-08-09 Reading time: 4 min ## In short - Buildings have thermal mass, and thermal mass is storage you have already paid for. - **Pre-cooling** runs the plant harder before the expensive window so the building coasts through it with reduced mechanical cooling. - It costs a small amount of extra energy and can produce a large reduction in peak demand. - How well it works depends on the building's construction — heavy structures hold a pre-cool, lightweight ones do not. - The strategy fails when the setpoints recover too fast at the end of the window and everything restarts at once. Cooling is the largest single component of the summer afternoon peak in most commercial buildings and in a great many industrial ones. It is also, uniquely among large loads, partly detachable in time — because the building itself stores the result. ## The mechanism Concrete, masonry, water in a loop and the contents of a building all hold heat. Cool them below the normal setpoint in the morning and the structure absorbs heat through the afternoon at a slower rate than the air alone would, which means less mechanical cooling is required during the hours when cooling is expensive. Nothing is installed. The store already exists, and until it is used it is simply the reason the building takes a while to warm up. How much store there is varies enormously and is worth measuring rather than guessing. The practical test is a free one: on a warm weekend, turn the cooling off and record how quickly the space temperature rises. A heavy building will drift slowly for hours; a lightweight one with a large glazed façade will be at outdoor conditions before lunch. That drift rate is, in effect, the discharge curve of the store you are proposing to use, and it tells you within a single weekend whether the strategy has anything to work with. ## The shape of the strategy Three phases, and each has a failure mode. **Pre-cool.** Beginning some hours before the peak window, drift the setpoint down, typically by one to two degrees. The plant runs harder while capacity is cheap. The pace matters: a slow drift is imperceptible, a step change is not. **Coast.** Through the expensive window, allow the setpoint to drift upward toward the top of the acceptable comfort range. Mechanical cooling reduces, and the structure gives back the coolth it absorbed. **Recover.** After the window, return to normal. **This is where the strategy most often fails**, because if every zone recovers at once, the plant restarts at full output simultaneously and produces an interval average that can exceed the peak the strategy just avoided. Recovery has to be staggered for exactly the reason a cold start does: [staggered startup](https://thedemandcharge.com/articles/staggered-startup-sequencing). ## What it is worth **Pre-cooling a building with a defined on-peak window** Cooling-driven summer peak. - Site peak, unmanaged: 940 kW - Chiller and air handling contribution at the peak interval: 385 kW - Reduction in mechanical cooling achieved during the coast phase: 45% - (Peak demand avoided: 385 × 0.45): 173 kW - All-in value of one avoided kW-month: $19.05 - Months in the cooling season: 5 - (Demand saving across the season): $16,478 - Additional energy consumed by pre-cooling, per season: 21,000 kWh - Energy rate, off-peak: 5.4 ¢/kWh - (Cost of the extra energy): $1,134 - **Net seasonal value: $15,344** *The extra energy is a real cost and it belongs in the model. It is also, on these figures, about seven percent of the benefit. Figures illustrative.* The ratio in that last line is why the strategy is attractive: the energy penalty is small and the demand benefit is large, provided the tariff prices demand meaningfully. ## Where it works and where it does not **Works well:** heavy construction with real thermal mass; a clearly defined and predictable peak window; a building management system capable of scheduled setpoint changes by zone; a comfort range with some latitude in it. **Works poorly:** lightweight construction with extensive glazing, where the pre-cool is lost within the hour; process cooling with a fixed temperature requirement and no tolerance; a tariff that measures facility demand across all hours, where the pre-cooling period may itself become the new peak. That last case is the one to check before implementing. Running the plant harder in the morning raises the morning load, and if the tariff bills a facility demand measured across every interval, the new morning peak may simply replace the afternoon one. Where the tariff bills only an on-peak window, the morning is free. The determinants have to be modeled separately: [facility, on-peak and billing demand](https://thedemandcharge.com/articles/facility-vs-on-peak-demand). ## Getting the window right The window is defined by the tariff, not by intuition, and the definition is more specific than most summaries suggest: start and end times, which days of the week, which holidays are excluded, and which months constitute the season. A schedule set thirty minutes wrong, or one that treats a holiday as a working day, produces peaks in the exact window it was written to avoid. Time-of-use schedules make the same window matter for energy pricing as well as demand, which improves the economics further: [time-of-use rates for commercial accounts](https://thedemandcharge.com/articles/time-of-use-rates-commercial). ## When to go further Pre-cooling uses a store you already own, which makes it cheap and also limits it. The store is as large as the building happens to be, it cannot be topped up on demand, and its behavior varies with weather. When the available shift is exhausted and a substantial cooling peak remains, the next step is a purpose-built store sized to the duty rather than to the architecture: [thermal energy storage for demand reduction](https://thedemandcharge.com/articles/thermal-energy-storage-demand). And where the remaining peak is not cooling at all, the strategy has run out of scope entirely and the ladder continues elsewhere: [how to reduce peak demand charges](https://thedemandcharge.com/articles/how-to-reduce-peak-demand-charges). ## Sources - [ASHRAE — Standards and guidelines](https://www.ashrae.org/technical-resources/standards-and-guidelines) - [ENERGY STAR for buildings and plants](https://www.energystar.gov/buildings) - [DOE Better Buildings Solution Center](https://betterbuildingssolutioncenter.energy.gov/) - [DOE Federal Energy Management Program](https://www.energy.gov/femp/federal-energy-management-program) ## Frequently asked questions ### Does pre-cooling use more energy? Usually slightly, yes. Cooling a building below its normal setpoint increases the temperature difference to outside and therefore the heat gain, and some of the stored coolth is lost. The trade is a small energy increase for a large demand reduction. ### Will occupants notice? They will if the setpoint swing is too wide or too fast. The strategies that work stay within the comfort range the building already operates in and move slowly. A drift of a degree or two over several hours is generally imperceptible; four degrees in thirty minutes is not. ### Does it work in every building? No. It depends on thermal mass. A heavy concrete structure holds a pre-cool for hours; a lightweight building with large glazing loses it quickly and may give back very little. ### Is this the same as thermal energy storage? Same principle, different medium. Pre-cooling uses the building's own structure as the store, which is free but limited and imprecise. A thermal storage tank is purpose-built, larger and controllable. --- Canonical URL: https://thedemandcharge.com/articles/hvac-scheduling-demand Site index for agents: https://thedemandcharge.com/llms.txt