HVAC Cooling Load Calculator

What the calculator computes

Enter the space and its design conditions, and the tool returns the sensible cooling load, the rate of heat the equipment has to remove to hold the room at its target temperature. It adds six heat gains: conduction through the walls and roof, solar radiation through the glass, the sensible heat from people, the heat from lighting, the heat from equipment, and the sensible load of the fresh air and infiltration. It then applies a safety margin and reports the result in kilowatts, tons of refrigeration, and BTU per hour, three units for the same number.

Alongside the headline figure the panel shows the load before the margin, the temperature difference that drives the envelope and ventilation gains, and the load intensity in watts per square meter. The chart breaks the load into its six parts so the biggest slice is obvious, because the way to a smaller unit is almost always to shrink the largest gain, not to guess at a bigger machine.

Cooling load as a sum of heat gains

A room heats up because energy flows into it from several directions at once. Heat conducts through the walls and roof whenever it is hotter outside than in. Sunlight pours through the windows as radiation. The people inside give off body heat, the lights convert electricity to heat, and every computer, motor, and appliance turns its wattage into warmth. On top of that, the fresh air brought in for ventilation, and the outside air that leaks in, arrives hot and has to be cooled. The cooling load is the sum of all these gains at the design condition.

Each gain is a rate, measured in watts, not a quantity of energy, so the load is the power the cooling system must deliver continuously to keep pace. Add the six rates, apply a margin for the things the simple model misses, and you have the design load. The method here is a steady-state heat-gain estimate: it assumes the peak gains happen together, which keeps the arithmetic transparent while still landing close to a fuller calculation for a typical space.

The envelope gain: U times A times dT

Heat crosses the walls and roof by conduction, and the rate depends on three things: how well the construction resists heat, how much surface area there is, and how big the temperature difference is. The formula is Q = U x A x dT. U is the overall heat transfer coefficient in watts per square meter per kelvin, a measure of how easily heat passes through the assembly, so a low U means a well-insulated wall. A is the area of wall and roof in square meters. dT is the outdoor design temperature minus the indoor target, in kelvin or degrees Celsius, which are the same size of step.

In the default space the envelope is 250 m2 at a U of 1.5 W/m2K, and the temperature difference is 35 minus 24, or 11 K. The envelope gain is 1.5 x 250 x 11 = 4,125 W. Lower the U by adding insulation and this gain falls in direct proportion, which is why the building fabric is the first lever in a hot climate. This simple model uses one combined U and area for the walls and the roof, though the roof often runs hotter under direct sun.

Solar gain through the glazing

Glass is transparent to sunlight, so radiation passes straight through a window and lands as heat inside. This solar gain is separate from the conduction through the glass and is usually far larger. The rate is the glazing area times a solar gain figure in watts per square meter, where that figure is the solar heat gain coefficient of the glass multiplied by the solar flux striking it. A clear single pane facing the sun can admit several hundred watts per square meter, while a coated low-SHGC unit with shading admits much less.

In the default space the glazing is 30 m2 at a solar gain of 250 W/m2, so the solar load is 30 x 250 = 7,500 W, larger than the entire envelope gain. That is the usual story in any space with a lot of glass: the window is the dominant heat source, and the cheapest path to a smaller cooling system runs through the glass, whether by shading it, specifying a lower SHGC, or having less of it on the sunny faces.

Internal gains from people, lighting, and equipment

Everything inside the room that gives off heat adds to the load. People are the first source: a seated adult sheds roughly 100 W of sensible heat, more with activity, and that figure times the number of occupants is the people gain. In the default office, 20 people at 100 W each add 2,000 W. This is the sensible part; people also give off moisture, a latent load handled separately, which this method does not size.

Lighting turns electricity into light and then into heat, so the lighting gain is the installed lighting power density in watts per square meter times the floor area. At 12 W/m2 across 200 m2 that is 2,400 W. Equipment is the third internal source, and here the rule is blunt: essentially all the electricity a device draws ends up as heat in the room, so you enter the installed load directly, the default 3,000 W covering the computers, screens, and small appliances of a busy office. Together the three internal gains add 2,000 + 2,400 + 3,000 = 7,400 W.

The ventilation and infiltration gain

A conditioned space needs fresh outdoor air for the people in it, and that air arrives at the outdoor temperature and has to be cooled to the indoor temperature. Outside air also leaks in through cracks and doors, which is infiltration. The sensible part of this load follows a compact formula: Q = 0.34 x volume x ACH x dT. The volume is the floor area times the ceiling height, in cubic meters. ACH is the air changes per hour, how many times the room volume is replaced with outside air each hour. The constant 0.34 is the volumetric heat capacity of air in watt-hours per cubic meter per kelvin.

In the default space the volume is 200 x 3 = 600 m3, the fresh-air rate is 1 air change per hour, and dT is 11 K, so the ventilation gain is 0.34 x 600 x 1 x 11 = 2,244 W. Raise the air change rate and this gain rises in step, which is why a high fresh-air requirement, common in crowded or regulated spaces, can become a significant slice of the load. This figure is the sensible ventilation load only. The fresh air also carries a latent load when it is humid, the energy to remove its moisture, which is not part of this sensible sum.

The safety margin and unit conversions

The six gains add up to the raw sensible load, but no simple model captures every detail, so a safety margin is added on top. A factor of about 10 percent covers small omissions, measurement slack, and a little future growth without grossly oversizing the machine. In the default space the six gains total 21,269 W, and a 10 percent margin lifts that to 23,396 W, which is 23.40 kW. The margin is a judgment, not a law: tighten it when the inputs are solid and loosen it when they are rough, but resist a large cushion, because an oversized unit brings its own problems.

The load can be read in three units. Kilowatts are the plain SI measure. A ton of refrigeration is the cooling rate that freezes one short ton of ice in a day, and it equals 3,517 W, or 3.517 kW, or 12,000 BTU per hour. So 23,396 W divided by 3,517 gives 6.65 tons. The British thermal unit per hour is common in North American equipment catalogs, and 1 kW is 3,412 BTU/hr, so the load is about 79,826 BTU/hr. The load intensity, 23,396 divided by the 200 m2 floor, is 117.0 W/m2, a figure you can compare against typical ranges to check the answer is sensible.

Reading the load breakdown and which gain dominates

The chart splits the load into its six parts, and the split is where the decisions live. For the default office the parts are the envelope at 4,125 W, the solar at 7,500 W, the people at 2,000 W, the lighting at 2,400 W, the equipment at 3,000 W, and the ventilation at 2,244 W. The solar gain is the single largest slice by a clear margin, which flags the glass as the thing to address first if the goal is a smaller system. In a different space the winner might be the equipment in a data closet or the ventilation in a packed meeting room, and the chart makes that obvious without any arithmetic.

Knowing the dominant gain changes what you do next. A big envelope slice points to insulation and a lower U, a big solar slice to shading or glass selection, a big internal slice to efficient lighting and equipment, and a big ventilation slice to heat recovery on the fresh air. Sizing a bigger unit treats the symptom; shrinking the largest gain treats the cause, and it lowers both the first cost of the equipment and the energy bill for its whole life.

Sensible versus latent load

Cooling does two jobs: it lowers the air temperature, the sensible load, and it removes moisture from the air, the latent load. This calculator sizes the sensible load only, the part that follows temperature difference. The latent load comes from the moisture that people give off, from the humidity in the fresh air, and from any wet process, and removing it takes energy at the coil even though it does not show as a temperature change until the water condenses.

The latent load is not small. In a humid climate or a space with a high fresh-air rate it can add 20 to 30 percent on top of the sensible figure, and in some cases more. A real equipment selection accounts for it through the coil’s sensible heat ratio and the dehumidification design. Treat the sensible load from this tool as the temperature-driven core of the job, and add a latent allowance from a psychrometric calculation before you commit to a coil, especially anywhere the air is damp.

Rule-of-thumb cross-checks

Fast rules of thumb are worth keeping as a sanity check, not as a design method. The most common one sizes cooling at a set floor area per ton, often quoted around 400 to 600 square feet per ton for an office, which is roughly 37 to 56 square meters per ton. The default space of 200 m2 at 6.65 tons works out to about 30 square meters per ton, a little denser than the rule because it carries a heavy solar and equipment load. When a result lands far outside a familiar rule, that is a prompt to check the inputs, not proof that either number is wrong.

The load intensity in watts per square meter is the same idea in SI units. A general office often falls between 80 and 150 W/m2, and the default result of 117.0 W/m2 sits comfortably inside that band. A result well under the band might mean a gain was left out; a result well over it might mean a double count or an input in the wrong unit. Rules of thumb cannot replace the component method, since they hide the very breakdown that tells you what to fix, but they catch gross errors before the equipment is bought.

Reading the results panel

The headline is the cooling load in kilowatts, the sensible heat the system must remove at the design condition. Directly under it, the tons and the BTU per hour restate that same load in the units equipment catalogs use, so you can match the number to a product without converting by hand. The load intensity gives the watts per square meter for a quick cross-check, and the before-margin figure shows the raw sum so you can see how much the safety factor added.

The temperature difference is shown because it drives both the envelope and the ventilation gains, and a wrong indoor or outdoor value moves the whole answer. The note names the dominant gain in plain words, and the chart sizes each of the six gains against the others. Read the panel top to bottom and you get the number to buy against, the units to buy in, and a clear target for cutting the load.

The limits of the method

This is a simplified steady-state estimate, and it earns its speed by leaving things out. It does not model the peak solar hour by orientation, so it cannot tell you that the east glass peaks in the morning and the west glass in the late afternoon, which a full calculation uses to avoid summing peaks that never happen at once. It does not model thermal mass, the way a heavy building soaks up heat and releases it later. And it does not track cloud cover, shading schedules, or the exact infiltration beyond the air change rate you enter.

It also sizes the sensible load only, so the latent load is absent, and in a humid climate that omission is large. The single combined U and area for walls and roof is a simplification, since the roof usually runs hotter under the sun. None of this makes the estimate useless: for a typical space it lands close. But it is a planning and teaching tool, so confirm the result with an ASHRAE 62.1 or 183 calculation, or an NBR 16401 study, and a mechanical engineer before buying equipment.

Where this calculator fits

It suits anyone who needs a defensible first number for the cooling of a space: a facilities manager scoping a replacement unit, an architect checking that a design will not overheat, a contractor sizing a quote, or a student working through a load-estimation exercise. The component breakdown makes it more than a single number, because it shows which gain to attack before a team commits to glass, insulation, and a machine size.

Because the tool reports in kW, tons, and BTU per hour at once, it bridges the SI and North American conventions in one screen. It is a screening and sizing aid that gets you to a credible load and a clear priority quickly, then hands off to a detailed design tool and an engineer for the final selection.

Common mistakes to avoid

The first mistake is forgetting the solar gain or lumping it into the envelope, which understates the load badly in any glassy space. The second is treating the sensible load as the whole job and skipping the latent load, which leaves a humid climate clammy even when the temperature is met. The third is confusing a ton of refrigeration with a metric tonne of weight or mishandling the BTU conversion, which throws the equipment size off by a wide margin.

A fourth is piling on a large safety factor on top of an already conservative model, which produces an oversized unit that short-cycles, controls humidity poorly, and costs more to buy and run. A fifth is entering an input in the wrong unit, area in square feet where the field asks square meters, or a lighting figure in total watts where it asks watts per square meter. Include every gain, keep the sensible and latent sides straight, convert units carefully, and size close to the real load, and the estimate will guide a sound decision.

Five worked examples

Example 1: the envelope and solar gains

Start with the default office of 200 m2 and a 3 m ceiling, at 35 C outside and 24 C inside, so the temperature difference is 11 K. The envelope of walls and roof is 250 m2 at a U of 1.5 W/m2K, so the conduction gain is U x A x dT = 1.5 x 250 x 11 = 4,125 W. The glazing is 30 m2 at a solar gain of 250 W/m2, so the solar gain is the glass area times the solar figure, 30 x 250 = 7,500 W. The glass alone admits more heat than the entire wall and roof assembly, which makes it the single largest gain in this space.

Example 2: the internal gains

Now count the heat generated inside the room. The 20 people each give off about 100 W of sensible heat, so the people gain is 20 x 100 = 2,000 W. The lighting runs at 12 W/m2 across the 200 m2 floor, so the lighting gain is 12 x 200 = 2,400 W. The equipment, the computers, screens, and small appliances, draws 3,000 W, and essentially all of it becomes heat, so the equipment gain is 3,000 W. Everything electrical in the room turns into a cooling load. The three internal gains add to 2,000 + 2,400 + 3,000 = 7,400 W, which rivals the solar gain and shows that a busy office heats itself nearly as much as the sun heats it.

Example 3: the ventilation gain

Fresh air for the occupants arrives hot and must be cooled. The room volume is the floor area times the ceiling height, 200 x 3 = 600 m3. At 1 air change per hour the whole volume is replaced once each hour with outside air at 35 C. The sensible ventilation load follows Q = 0.34 x volume x ACH x dT = 0.34 x 600 x 1 x 11 = 2,244 W, where the constant 0.34 is the volumetric heat of air in watt-hours per cubic meter per kelvin. If the space needed 2 air changes per hour for a denser crowd, this gain would double to 4,488 W, which is why ventilation-heavy rooms carry a load out of proportion to their floor area.

Example 4: the total and the margin

Add the six gains together: envelope 4,125, solar 7,500, people 2,000, lighting 2,400, equipment 3,000, and ventilation 2,244, which sum to 4,125 + 7,500 + 2,000 + 2,400 + 3,000 + 2,244 = 21,269 W. That is the raw sensible load, 21.27 kW before any allowance. A 10 percent safety margin covers the small things the simple model leaves out and a little growth, and it lifts the load to 21,269 x 1.10 = 23,396 W, which is 23.40 kW. This is the figure to size the equipment against. The margin is deliberate and modest; a larger cushion would only push the selection toward an oversized machine that runs poorly at part load.

Example 5: into tons and BTU/hr

Convert the final load into the units equipment is sold in. One ton of refrigeration is 3,517 W, so 23,396 divided by 3,517 is 6.65 tons. To get BTU per hour, multiply the kilowatts by 3,412, so 23.396 x 3,412 is about 79,826 BTU/hr, which also equals 6.65 tons times 12,000. The load intensity is the final load divided by the floor area, 23,396 / 200 = 117.0 W/m2, in the normal band for a glassy office. So the same space reads as 23.40 kW, 6.65 tons, and 79,826 BTU/hr, three names for one cooling job.

Three expert tips

Attack the biggest slice first

The breakdown exists so you can find the largest gain and cut it, rather than reaching for a bigger machine. In a glassy space that biggest slice is almost always the solar gain, as in the default office where the glass admits 7,500 W, more than the walls and roof combined. Shading the windows, specifying a lower solar heat gain coefficient, or simply putting less glass on the west and south faces cuts more load than any increase in chiller size, and it does so for the whole life of the building rather than just at purchase. Size the equipment to the load you actually need after the biggest gain is tamed, not to the load of an unshaded design.

Do not confuse sensible and latent load

This method sizes the sensible heat only, the part tied to temperature difference, and it stops there on purpose. A humid climate or a high fresh-air rate adds a latent load, the energy to remove moisture from the air, and that latent load can be 20 to 30 percent more on top of the sensible figure, sometimes more. It is handled by the cooling coil and the dehumidification design, through the sensible heat ratio the coil is selected for. If you size a machine on the sensible load alone in a damp place, the room will hit its temperature but feel clammy, so add a latent allowance from a psychrometric calculation before you choose the coil.

Size close to the real load

There is a standing temptation to round the tonnage way up for peace of mind, but a generous cushion backfires. An oversized unit reaches the set temperature quickly and then shuts off, so it short-cycles, switching on and off too often, which wears the compressor and wastes energy. It also runs for too little time to pull much moisture out of the air, so it controls humidity poorly, leaving the space cool but damp. And it costs more to buy and more to run at its inefficient part load. Size close to the real load, keep the safety margin modest at around 10 percent, and let a right-sized machine run longer, steadier cycles that hold both temperature and humidity.

Frequently asked questions

Sources, disclaimer, and editorial transparency

The heat-gain formulas, the envelope conduction, the solar gain through glazing, the internal gains, the ventilation load, the unit conversions, and the limits described here follow recognized HVAC engineering sources, including the Engineering ToolBox cooling load equations and standard cooling-load references such as the CED Engineering ASHRAE cooling-load design courses. Loads are computed as a simplified steady-state sum of sensible heat gains with a safety margin, and a full ASHRAE or NBR 16401 calculation is recommended before purchasing equipment. This calculator and guide are built and reviewed by the OpsCalculators team; see our Editorial Policy for how each tool is researched, built, and tested.

Results are accurate estimates for planning and education, not a substitute for a full mechanical design or a site survey. The method sizes the sensible load only and leaves out the latent load, the peak solar hour by orientation, and thermal mass, so validate outputs before a capital or equipment decision. See our full Disclaimer. OpsCalculators.com is operated by MAFHH INTERNATIONAL LTD. Your inputs are processed in your browser and are never stored; see our Privacy Policy.