A load calculation answers one question: how many Btu per hour must the equipment add in winter, and take out in summer, to hold the indoor temperature on a design day? This page lists every step the calculator takes, so you can check it or redo it by hand. It is an open first-principles method. It is not ACCA Manual J, whose tables and procedure belong to ACCA, and it simplifies in the ways listed at the end.
A wood-framed wall is two heat paths side by side: through the insulated cavity and through the studs. Each path's resistance is the sum of its layers, and the wall's U-factor is the area-weighted sum of the two paths' conductances (the parallel-path method):
U = (1 − f) / R_cavity_path + f / R_stud_path
With inside air film R 0.68, ½" drywall 0.45, sheathing 0.62, siding 0.62, outside film 0.17, softwood at R 1.25 per inch and a framing fraction f of 0.25, a 2x4 wall with R-13 batts works out at U ≈ 0.084 Btu/h·ft²·°F, about R-11.8 overall rather than the R-13 on the bag. The attic floor is 1 / (R_insulation + 0.45 + 2 × 0.61). Windows and doors use the U-factor from their label. A slab loses heat mostly at its edge, so it is counted as F × perimeter × ΔT; the default F of 0.73 Btu/h·ft·°F is the usual figure for an uninsulated slab edge, and an insulated edge is lower.
Heating is steady state at the winter design temperature, with no credit for sun or people, since the worst hour is a cloudy pre-dawn:
Q_heat = Σ U·A·ΔT + F·P·ΔT + 1.08 · ρ · cfm_leak · ΔT
Air leakage comes from a blower-door number. A house's natural air change rate is roughly ACH50 divided by an N factor (about 20 for a typical single-storey house; lower for tall or windy sites), so 6 ACH50 is 0.3 air changes an hour. That airflow is shared among rooms by their outside wall area. 1.08 is the sensible heat of air (Btu/h per cfm per °F) at sea level; ρ scales it down with altitude by the standard atmosphere (about 83% at 5,000 ft). Mechanical ventilation adds 1.08 · ρ · cfm · (1 − recovery) · ΔT to the whole house, and ducts outside the conditioned space add a percentage on top.
Summer loads swing with the sun, so the calculator runs 24 solar hours of a clear 21 July:
A · SHGC · shade · (beam · K(θ) + 0.9 · diffuse), where K(θ) trims the beam at steep incidence, plus conduction U · A · (T_out − T_in).T_out + α·I / 3.0, with α 0.45, 0.7 or 0.9 for light, medium or dark). A frame wall passes 80% of the sol-air swing about its daily mean through to the inside, two hours late; masonry passes 35%, six hours late.atticRiseF).0.68 · ρ · cfm · Δgrains, with outdoor humidity from the design dew point and indoor from the set point and RH (75 °F at 50% is about 65 grains per pound).An east bedroom peaks around 9 am, a west living room near 4 pm. The equipment sees the hour when the house total is largest, which is smaller than adding every room's worst hour. The calculator reports both: the house peak sizes the equipment, and each room's own peak sizes its duct. Design airflow is sensible load / (1.08 · ρ · supply ΔT), shared among rooms by the larger of each room's heating and cooling share.
The page loads an illustrative 1,800 ft² single-storey ranch on a slab (36°N, 600 ft, 22 °F winter and 93 °F summer design, 72 °F dew point, R-13 walls, R-38 attic, U-0.32 / SHGC-0.27 windows, 6 ACH50, 50 cfm ventilation, ducts in the attic at 10%). It is not a measured house. The result:
| Cooling, at the house peak (3 pm solar) | Btu/h |
|---|---|
| People and appliances | 4,400 |
| Sun through windows | 3,946 |
| Walls | 3,470 |
| Ceiling to attic | 2,222 |
| Ducts in the attic | 1,794 |
| Air leakage | 1,434 |
| Window conduction | 1,083 |
| Ventilation | 951 |
| Doors | 432 |
| Sensible total | 19,732 |
| Latent (moisture) | 6,569 |
| Total | 26,301 (2.19 tons) |
The heating load is 26,901 Btu/h. The 1-ton-per-500-ft² rule of thumb would have put a 3.5 to 4 ton unit in this house; the load says a 2.5 ton unit covers it, at about 820 ft² per ton. That gap is the reason to run the numbers: an oversized system costs more, short-cycles and pulls less moisture out of the air. Adding up each room's own peak gives 17,910 Btu/h sensible, against 16,987 for the same rooms at the house's 3 pm peak (both before ventilation and ducts), a 5% gap: most of this house's glass faces south and west, so its rooms peak within a few hours of each other. A house with large east glass would show a bigger gap.
Give a wall an overhang depth, the gap from the window head up to the overhang and the window height, and each hour the calculator works out where the shadow edge falls. It drops below the overhang by depth × tan(altitude) ÷ cos(sun azimuth − wall azimuth), the sun's height seen square-on to the wall. The shaded share of the window's height gets no direct sun; the sky and ground light are kept in full, which slightly overstates the load. The overhang is assumed to run well past both sides of the window.
On the example house a 2 ft eave 1 ft above every 5 ft window lowers the 3 pm sensible peak by 3.3% (19,732 to 19,072 Btu/h). The south rooms drop 7 to 13%, because the high midday sun is what an eave stops; west glass at 3 pm barely changes, because that sun is low enough to get under it.
Pick "Basement (below grade)" as a room's floor, then give how deep the floor sits below grade, the length of wall below grade and the basement's shortest width. Walls above grade go in the room's walls list as usual, with their above-grade height. Heat leaves a below-grade wall along roughly circular paths through the soil to the ground surface, so the deeper a strip of wall, the longer its path. Averaged from grade down to depth z, the wall's U-factor is 2k ÷ (π z) × ln((z + 2kR/π) ÷ (2kR/π)), and the floor's, for shortest width w, is 2k ÷ (π w) × ln((w/2 + z/2 + kR/π) ÷ (z/2 + kR/π)): the path-length method of Latta and Boileau as given in ASHRAE Fundamentals, with k the soil conductivity and R the concrete, insulation and inside film. Both work against the ground surface design temperature, the annual mean outdoor temperature less its seasonal swing, not the winter air temperature. In summer the ground is cooler than the room, so below-grade gains are taken as zero.
Checked against a 2-D soil model solved cell by cell (eval/basement-fd.mjs): the wall formula reads 10 to 16% low on bare concrete and within 5% with R-10 on it; the floor formula reads 5 to 43% low against the same model, and between 31% low and 24% high when the deep ground is held at the annual mean. On a cold-climate example the floor is about 4% of the house's heating load, so that gap is about 1% of the total.
Each room can carry skylights with an area, a tilt (0 for flat, 18.4° for a 4-in-12 roof) and the direction the roof slopes toward. Winter: area × skylight U × the temperature difference. Summer: conduction plus the sun on glass at that tilt and facing, hour by hour, through the skylight SHGC with the same angle modifier as windows; the skylight's area comes out of the attic ceiling. On the example ranch, two 8 ft² skylights add 1,260 to 1,410 Btu/h of sun at the house peak (about 0.13 tons), more when the roof faces west.
For a permit, or before buying equipment, have a full load calculation done with approved software. Use this one to know whether the number on a quote is in the right range, and to ask why if it is not.
Because walls, windows and the attic peak at different times. A west window peaks late in the afternoon, an east one in the morning, and a wall's heat arrives a couple of hours after the sun hits it. Running 24 hours finds the hour when the whole house needs the most cooling.
You enter them. Use the 99% heating and 1% cooling design values for your county, which your local code office, an energy-code compliance tool or a weather-data source lists. Do not use record extremes; sizing for them oversizes the equipment.
Usually not. Many building departments require a load calculation from approved software. This page shows the method so you can check whether a quote or report is in the right range.