- A load calculation adds up every way heat gets in or out: walls, the attic, glass, sun, air leaks, people, appliances and ducts. Square feet alone cannot tell you that.
- On a typical house the sun through east and west glass is the biggest cooling load. Low-E glass cut the sample house's cooling load by about 40%.
- Manual S lets an air conditioner be no more than 115% of the total cooling load. A furnace can be 100% to 140% of the heating load.
What a Manual J load calculation is
A load calculation works out how many Btu per hour a house gains on a hot design afternoon and loses on a cold design night. ACCA Manual J is the standard method in the US, and most building codes point to it before a new system goes in. It is done room by room with software, using the house's real wall, window and duct details.
This calculator uses the same physics on the whole house at once. It is an estimate for planning and quoting, not a permit-grade Manual J. When the building department asks for a load report, run the full calculation. When you need to check a bid, spot an oversized replacement, or explain to a homeowner why the old 5-ton unit short cycles, this gets you close and shows its work.
The load has two parts. Sensible heat changes the air temperature: heat through walls, the attic and glass, sun through windows, warm outdoor air leaking in, people and appliances. Latent heat is moisture the system has to pull out of the air. Add them and divide by 12,000 to get tons.
How the calculator works out each part
Every part is a short formula. These are the numbers behind each one.
| Part | Formula | Where the numbers come from |
|---|---|---|
| Walls | U × net wall area × (outdoor − 75°F) | 2021 IECC U-factors for code-level walls, typical values for older ones |
| Attic ceiling | U × ceiling area × (attic − 75°F) | FSEC measured attic temperatures by roof type |
| Glass, conduction | U × glass area × (outdoor − 75°F) | Typical glass values, the code low-E limit, or your window sticker |
| Sun through glass | Glass area × solar heat gain factor × (SHGC ÷ 0.87) | ASHRAE July factors at 40° north, worst hour from 8 am to 4 pm |
| Air leakage | 1.08 × cfm × temperature difference | Blower door ACH50 ÷ 20 for natural air changes (rule of thumb) |
| People | 230 Btu/h sensible, 200 latent each | Manual J, counting bedrooms plus one |
| Ducts outside | +20% of the sensible load | Low end of ENERGY STAR's 20 to 30% duct leakage |
| Moisture | 0.68 × cfm × grains of moisture difference | Outdoor humidity from the ASHRAE design wet bulb |
Inside conditions are the usual Manual J ones: 75°F at 50% humidity for cooling and 70°F for heating. Outdoor conditions are the ASHRAE 1% cooling and 99% heating temperatures for the city you pick. Each city is the Department of Energy's sample city for its climate zone. If you are somewhere else, choose "Somewhere else" and enter your own design temperatures from Manual J Table 1A.
| City (climate zone) | Summer 1% | Wet bulb | Winter 99% |
|---|---|---|---|
| Miami (1A) | 90.9°F | 77.5°F | 52.0°F |
| Houston (2A) | 94.5°F | 78.3°F | 36.0°F |
| Phoenix (2B) | 108.3°F | 69.4°F | 41.5°F |
| Atlanta (3A) | 91.6°F | 73.9°F | 26.4°F |
| Las Vegas (3B) | 106.3°F | 67.1°F | 33.8°F |
| Baltimore (4A) | 91.2°F | 74.1°F | 18.0°F |
| Seattle (4C) | 81.7°F | 63.7°F | 29.7°F |
| Chicago (5A) | 88.7°F | 73.2°F | 3.7°F |
| Denver (5B) | 91.8°F | 59.9°F | 6.6°F |
| Minneapolis (6A) | 88.0°F | 72.0°F | -6.2°F |
The wet bulb matters as much as the dry bulb. Houston and Phoenix both run hot, but Houston's outdoor air carries far more moisture. In the calculator that shows up as a latent load in Houston and almost none in Phoenix.
Worked example: a 2,000 sq ft house in Atlanta
The calculator opens on this house: 2,000 sq ft on one floor with 8 ft ceilings, R-13 walls and an R-30 attic under shingles with soffit and ridge vents, 230 sq ft of double-pane clear glass (60 north, 50 east, 70 south, 50 west), average air tightness of 7 ACH50, four people, and ducts in the attic.
| Part | Math | Cooling |
|---|---|---|
| Walls | 0.084 × 1,201 sq ft × 16.6°F | 1,675 Btu/h |
| Attic ceiling | 0.035 × 2,000 × (113.6 − 75)°F | 2,702 Btu/h |
| Glass, conduction | 0.48 × 230 × 16.6°F | 1,833 Btu/h |
| Sun through glass | Worst hour 10 am, SHGC 0.76 | 14,632 Btu/h |
| Air leakage | 93 cfm × 1.08 × 16.6°F | 1,673 Btu/h |
| People | 4 × 230 | 920 Btu/h |
| Ducts in the attic | 20% of 23,435 | 4,687 Btu/h |
| Moisture | 0.68 × 93 cfm × 32.6 grains + 4 × 200 | 2,867 Btu/h |
| Total | 30,989 Btu/h (2.58 tons) |
The heating load is 19,991 Btu/h at Atlanta's 26.4°F design temperature. The house works out to about 774 sq ft per ton, so the old "500 sq ft per ton" rule would have put a 4-ton unit on a 2.6-ton house.
Look at where the cooling goes. The sun through glass is almost half of it. The walls are under 6%. That is why window direction and glass type move the answer more than anything else you put in.
| Change to the same house | Cooling load | Heating load |
|---|---|---|
| As above | 30,989 Btu/h | 19,991 Btu/h |
| Ducts inside the living space | 26,302 Btu/h | 16,659 Btu/h |
| Low-E glass (U-0.30, SHGC 0.25) | 18,382 Btu/h | 17,825 Btu/h |
| Soffit vents only, no ridge vent | 32,081 Btu/h | 19,991 Btu/h |
| Leaky (15 ACH50) | 35,646 Btu/h | 26,018 Btu/h |
| Low-E, tight, code insulation, ducts inside | 12,485 Btu/h | 10,301 Btu/h |
| Uninsulated, single pane, leaky | 49,049 Btu/h | 49,050 Btu/h |
The same 2,000 sq ft runs from about 1 ton to about 4 tons depending on how it is built. Square feet alone cannot tell you which.
Getting the inputs right
Windows by direction. Measure the glass on each side and add it up. East and west glass take the low morning and afternoon sun straight on. North glass gets almost no direct sun. The calculator checks five times of day and keeps the worst one, because east glass peaks in the morning and west glass in the afternoon, never both at once.
Insulation. The top two levels are the 2021 IECC maximum U-factors: "Good" is the climate zone 3 level (U-0.060 walls, U-0.026 attic) and "Best" the zone 5 and 6 level (U-0.045, U-0.024). The lower two use typical values: "Some" is R-13 walls (about U-0.084 once the studs are counted) under an R-30 attic (about U-0.035); "Little or none" is an uninsulated 2x4 wall (about R-3.7 for the whole wall) under R-11 in the attic.
Roof. The attic temperature comes from Florida Solar Energy Center measurements in 21 homes on design afternoons:
| Roof over the attic | Attic on a design afternoon |
|---|---|
| Shingles, soffit vents only | Outdoor + 35°F |
| Shingles with a radiant barrier | Outdoor + 25°F |
| Shingles, soffit and ridge vents | Outdoor + 22°F |
| Tile | Outdoor + 10°F |
| White or reflective | Outdoor − 1.5°F |
Air tightness. If the house has had a blower door test, choose "I have a blower door number" and enter the ACH50. New homes built to recent energy codes often test at 3 ACH50 or tighter. Older, drafty homes can test at 15 or more. The calculator divides by 20 to get natural air changes, a common rule of thumb.
People. Manual J counts bedrooms plus one, not who lives there today, so the system is sized for the house rather than the current family.
Fresh air. If the house has a ventilation system, the calculator adds HVI's continuous rate: 5 cfm per 100 sq ft of floor, about 0.35 air changes per hour.
From the load to the equipment size
ACCA Manual S sets how far equipment can stray from the load. An air conditioner or heat pump should cover the total cooling load and be no more than 115% of it. A furnace should be between 100% and 140% of the heating load.
The calculator picks the smallest standard size (1.5 to 5 tons in half-ton steps) inside that band. When no size fits, as with the Atlanta house at 2.58 tons where 3 tons would be 116%, it shows the size just under and says so. Manual S allows a unit a little under the total load when its sensible capacity still covers the sensible load. Check the maker's expanded performance data at your design temperatures before you commit.
Oversizing is the common mistake. An oversized air conditioner cools the air fast and shuts off before it has pulled out much moisture, so the house feels cold and clammy. It also short cycles, which wears parts, and costs more to buy. When a load comes in well under the old system, that is usually the old system being oversized, not the load being wrong.
The sensible heat ratio in the results tells you how much of the load is temperature and how much is moisture. Humid climates have lower ratios. Pick equipment whose sensible capacity at your conditions covers the sensible load, not only the total.
What this estimate leaves out
- It works on the whole house, not room by room. Room loads set the airflow to each register, so they are needed to design ducts.
- Sun angles are for 40° north in July. Glass in Miami or Minneapolis gets a somewhat different sun.
- It treats the footprint as a square. A long or L-shaped house has more wall, so a bit more load.
- No shading from overhangs, trees or blinds, and no sun on walls. Both would lower the cooling load a little.
- Slab edge and basement wall losses are not counted in heating.
Once you know the size, build the estimate in Koira with good, better and best equipment options, and let the homeowner pick and sign from their phone. Need the ducts next? Use the duct size calculator.
Last checked October 8, 2026. More for your trade: how HVAC companies run on Koira.