Engineering Guide

HVAC Load Calculation Guide: Manual J, Sensible & Latent Heat Explained

Published June 12, 2026 · by Industrial Unit Converter Editorial Team

Why load calculations are the foundation of every HVAC design

Every HVAC system — whether a single split system for a 200 ft² office or a 500-ton chiller plant for a hospital — starts with one number: the calculated cooling and heating load. Get it wrong at the start and you'll pay for it for the life of the equipment.

Undersizing consequences:

Can't hold temperature setpoint on design days
Occupants uncomfortable, productivity drops
Equipment runs continuously, shortens service life
May fail code-required temperature and humidity limits

Oversizing consequences:

Short-cycles (frequent on/off), leading to compressor failure
Poor humidity control — the unit satisfies temperature before it runs long enough to dehumidify
Higher first cost for unnecessarily large equipment
Lower part-load efficiency (most equipment peaks at 50-80% of full load)
Larger electrical service, larger ducts, larger everything

The gold standard: Energy codes including ASHRAE 90.1 generally recommend that installed cooling capacity not exceed calculated peak load by more than 15%. Anything beyond that margin is likely waste.

This guide covers the methodology, the math, and a fully worked example you can use to cross-check your own calculations.

The three load calculation methodologies

Manual J (Residential)

Developed by ACCA (Air Conditioning Contractors of America), Manual J is the industry standard for single-family residential and small commercial buildings up to ~25,000 ft². It uses a room-by-room approach:

Calculate heat gain/loss for each room through walls, windows, roof, floor, infiltration
Sum for each zone
Size equipment for the zone peak (not the sum of room peaks — diversity matters)

Manual J uses simplified factors (HTM — Heat Transfer Multipliers) that pre-calculate the U×A×ΔT term for common construction types, reducing the calculation to: Load = HTM × Area for each surface.

ASHRAE RTS (Radiant Time Series) — Commercial

For larger commercial buildings, steady-state methods like Manual J become inaccurate because they ignore thermal mass time delay. Sun hitting a west-facing window at 4 PM generates heat that doesn't appear as cooling load until 6-7 PM, after the concrete floor and interior walls have absorbed and re-radiated it.

RTS accounts for this by splitting heat gains into:

Convective portion — immediately becomes cooling load (40-60% for most surfaces)
Radiant portion — absorbed by building mass, released later according to Radiant Time Factors

The total cooling load at hour H = Σ (convective gain at H + radiant gain from previous hours × their respective RTFs).

EnergyPlus / DOE-2 — Large and complex buildings

Full dynamic simulation using hourly weather data. Used for buildings over 100,000 ft², LEED certification, and anywhere energy code compliance modeling is required. Not covered in this guide, but the underlying physics are the same as RTS.

The load calculation equation: breaking down every component

The total cooling load for a space is the sum of:

Q_total = Q_envelope + Q_internal + Q_ventilation + Q_infiltration

1. Envelope loads (Q_envelope)

Heat entering through the building skin:

Q_wall = U × A × CLTD (or ΔT for steady-state)

Where:

U = overall heat transfer coefficient (BTU/hr·ft²·°F) — lower is better
A = surface area (ft²)
CLTD = Cooling Load Temperature Difference (°F) — accounts for time lag and solar effects

Construction Type	Typical U-Value (BTU/hr·ft²·°F)
Uninsulated 8" concrete block	0.50-0.80
Insulated 8" concrete block (R-10)	0.08-0.12
2×4 wood stud, R-13 batt	0.08-0.10
2×6 wood stud, R-19 batt	0.06-0.07
Metal curtain wall, spandrel	0.10-0.15
Single-pane window	0.90-1.10
Double-pane low-E window	0.30-0.40
Triple-pane low-E window	0.15-0.25

For windows, also include solar heat gain: Q_solar = A × SHGC × Solar_Intensity where SHGC (Solar Heat Gain Coefficient) ranges from 0.25 to 0.80.

2. Internal loads (Q_internal)

People: Sensible ~245 BTU/hr per person (seated office work), latent ~205 BTU/hr per person. Total ~450 BTU/hr per person for office occupancy.

Lighting: Q_lighting = Watts × 3.412 BTU/hr per watt × use_factor. For LED office lighting at 0.7 W/ft²: 0.7 × 3.412 = 2.39 BTU/hr·ft².

Equipment: Q_equip = Watts × 3.412 × use_factor. A typical office with computers and monitors: 1.5-2.5 W/ft² = 5-8.5 BTU/hr·ft².

3. Ventilation load (Q_ventilation)

Outside air brought in for indoor air quality, per ASHRAE 62.1:

Q_vent = 4.5 × CFM_OA × (h_OA − h_RA) (total, BTU/hr)

Or split into sensible and latent:

Q_vent_sensible = 1.08 × CFM_OA × (T_OA − T_RA) Q_vent_latent = 0.68 × CFM_OA × (W_OA − W_RA) (grains/lb)

Ventilation requirements per ASHRAE 62.1:

Offices: 5 CFM/person + 0.06 CFM/ft²
Classrooms: 10 CFM/person + 0.12 CFM/ft²
Conference rooms: 5 CFM/person + 0.06 CFM/ft²

4. Infiltration load (Q_infiltration)

Uncontrolled air leakage through the building envelope:

Q_infiltration = 1.08 × CFM_inf × (T_OA − T_RA)

CFM_inf can be estimated by blower door test (residential) or by the crack method: CFM_inf = (ACH × volume) ÷ 60, where natural infiltration ACH ranges from 0.1 (tight building) to 0.5 (leaky) in commercial construction.

Sensible vs latent: the split that defines every coil selection

Total cooling = Sensible + Latent. Always. Every cooling coil does both. The ratio determines the coil leaving air condition, the supply airflow, and whether the space will be comfortable.

Climate	Typical SHR (Sensible Heat Ratio)	Implication
Phoenix — hot, dry	0.85-0.95	Mostly sensible; small coil, higher airflow
Chicago — mixed	0.70-0.80	Balanced sensible and latent
Miami — hot, humid	0.55-0.70	High latent load; deep coil, lower airflow
Singapore — tropical	0.50-0.65	Latent-dominated design

Why this matters: A coil selected for SHR = 0.75 (typical) installed in a space with actual SHR = 0.55 will satisfy temperature but leave relative humidity at 65-70% — violating ASHRAE 55 comfort. The fix is either a deeper coil (more rows), lower airflow (600-700 fpm face velocity), or a dedicated outdoor air system (DOAS) that handles latent load separately from sensible.

Fully worked example: 5,000 ft² commercial office in Chicago

Building description

5,000 ft² open-plan office, 10 ft ceiling height (50,000 ft³)
40' × 25' south-facing exterior wall, 40% glazing (400 ft² windows)
R-13 insulated 2×4 stud wall (U = 0.09)
Double-pane low-E windows (U = 0.35, SHGC = 0.30)
Flat roof, R-30 insulation (U = 0.03)
50 occupants (100 ft²/person)
LED lighting: 0.7 W/ft²
Equipment: 2.0 W/ft²
Design conditions: Chicago 1% cooling design — 91°F DB, 74°F WB per ASHRAE Fundamentals
Indoor design: 75°F DB, 50% RH max

Step 1: Envelope loads

South wall (net of windows): Area = (40 × 10) − 400 = 0 ft² (all glazed — wall area behind windows is negligible in this orientation)

Actually, let me correct: the south wall is 40 ft × 10 ft = 400 ft² total. 40% glazing = 160 ft² windows, 240 ft² opaque wall.

Q_wall = 0.09 × 240 × (95 − 75) = 0.09 × 240 × 20 = 432 BTU/hr (assuming CLTD ≈ 95°F sol-air temperature)

South windows (conduction): Q_window_cond = 0.35 × 160 × (91 − 75) = 0.35 × 160 × 16 = 896 BTU/hr

South windows (solar): Q_window_solar = 160 × 0.30 × 150 = 7,200 BTU/hr (peak solar ~150 BTU/hr·ft² for south-facing at 4 PM in July at Chicago latitude)

Roof: Q_roof = 0.03 × 5,000 × 40 = 6,000 BTU/hr (CLTD ≈ 40°F for dark roof on design day)

Total envelope: 432 + 896 + 7,200 + 6,000 = 14,528 BTU/hr

Step 2: Internal loads

People: Q_people_sensible = 50 × 245 = 12,250 BTU/hr Q_people_latent = 50 × 205 = 10,250 BTU/hr

Lighting: Q_lighting = 5,000 × 0.7 × 3.412 = 11,942 BTU/hr

Equipment: Q_equip = 5,000 × 2.0 × 3.412 × 0.8 (diversity) = 27,296 BTU/hr

Total internal: 12,250 + 11,942 + 27,296 = 51,488 BTU/hr sensible + 10,250 BTU/hr latent

Step 3: Ventilation load

Per ASHRAE 62.1: 50 people × 5 CFM/person + 5,000 ft² × 0.06 CFM/ft² = 250 + 300 = 550 CFM OA

Sensible: Q_vent_sens = 1.08 × 550 × (91 − 75) = 1.08 × 550 × 16 = 9,504 BTU/hr

Latent: At 91°F/74°F WB, W_OA ≈ 100 gr/lb. At 75°F/50% RH, W_RA ≈ 65 gr/lb. ΔW = 35 gr/lb.

Q_vent_lat = 0.68 × 550 × 35 = 13,090 BTU/hr

Step 4: Total load

Component	Sensible (BTU/hr)	Latent (BTU/hr)
Envelope (wall, windows, roof)	14,528	0
People	12,250	10,250
Lighting	11,942	0
Equipment	27,296	0
Ventilation (OA)	9,504	13,090
Subtotal	75,520	23,340
Safety factor (10%)	7,552	2,334
Total	83,072	25,674

Grand total = 83,072 + 25,674 = 108,746 BTU/hr = 9.1 tons (at 12,000 BTU/hr/ton)

Sensible Heat Ratio = 83,072 ÷ 108,746 = 0.76

Step 5: Select equipment

At 400 CFM/ton: 9.1 × 400 = 3,640 CFM supply air

At SHR = 0.76, the coil must be selected for this sensible/latent split. Select a 10-ton packaged rooftop unit with a 6-row DX coil rated for SHR ≤ 0.75 at design airflow.

Converting load to equipment tonnage

The raw load calculation produces BTU/hr. Equipment is selected in tons (of refrigeration). The conversion is straightforward:

1 ton of refrigeration = 12,000 BTU/hr

So: Tons = Total Load (BTU/hr) ÷ 12,000

For our worked example: 108,746 ÷ 12,000 = 9.1 tons → select a 10-ton unit.

Typical tonnage benchmarks by building type

These are approximate ranges for preliminary planning only — never substitute for a proper load calculation:

Building Type	Typical ft² per Ton	Notes
Residential (older, uninsulated)	300-400	Higher envelope loads
Residential (new, code-compliant)	500-700	Better insulation, tighter envelope
Office — interior zone	250-350	Internal loads dominate
Office — perimeter zone	200-300	Envelope + internal loads
Retail — big box	250-400	High ceiling, high lighting load
Data center / server room	50-100	Equipment load dominates (100-200 W/ft²)
Classroom	200-300	High occupancy density
Restaurant	80-150	Kitchen equipment + high ventilation
Hospital patient room	200-300	High ventilation, 24/7 operation
Light manufacturing	150-250	Process load dependent

Warning: The "ft² per ton" rule has killed more HVAC designs than any other single error. A south-facing glass conference room in Phoenix (150 ft²/ton) and an interior break room in Seattle (800 ft²/ton) are in completely different universes. Use benchmarks only to sanity-check a complete calculation.

How to determine what ton air conditioner you need

Perform a proper load calculation (Manual J for residential, RTS for commercial)
Convert the total load to tons: Tons = BTU/hr ÷ 12,000
Select the next standard equipment size that covers the load
Verify the sensible/latent split — the unit must handle both
Check the sensible heat ratio of the selected equipment against your calculated SHR

Standard packaged equipment sizes: 1.5, 2, 2.5, 3, 3.5, 4, 5, 6, 7.5, 8.5, 10, 12.5, 15, 20, 25, 30, 40, 50 tons. For our 9.1-ton example, select a 10-ton unit.

Use our HVAC Load Calculator for quick estimates and our BTU/hr to kW Converter for metric conversions. The CFM per Ton Calculator verifies that your selected equipment delivers adequate airflow.

Common load calculation pitfalls

1. Ignoring thermal mass time delay

Steady-state methods (like simple U×A×ΔT) overestimate peak cooling load by 15-25% compared to RTS for medium-weight buildings because they assume all solar gain becomes instantaneous cooling load. The time delay matters — sunlight entering at 3 PM may not become load until 6 PM when the mass has stored and re-released it.

2. Using the wrong design conditions

ASHRAE design conditions are statistical — the 0.4% or 1% annual cumulative frequency of occurrence, not the all-time record. Using a Design Day that's 8°F hotter than the ASHRAE 0.4% condition doesn't add safety — it adds cost with no benefit. Trust the statistical design data.

3. Adding too many safety factors

"If the calculated load is 8 tons, I'll spec a 10-ton unit just to be safe." That's a 25% oversize. Now add the engineer's safety factor (10%) and the contractor's (another 10%) — the installed unit is 50% larger than needed. Safety factors compound. Apply only one margin, and do it at the end.

4. Forgetting the latent load

The most common error in commercial load calcs: calculating sensible load carefully, then selecting equipment based on total BTUH without verifying the sensible/latent split. A rooftop unit with a SHR of 0.80 can't handle a space with SHR of 0.65 — you'll hit temperature but never control humidity. Always check the unit's SHR rating against the calculated SHR.

5. Using "500 ft² per ton" or other rules of thumb

500 ft²/ton, 400 ft²/ton, 300 ft²/ton — none of these rules account for building orientation, window-to-wall ratio, climate zone, internal load density, or ventilation requirements. A south-facing all-glass office in Phoenix is a fundamentally different load from a north-facing interior space in Seattle. Rules of thumb should only be used to sanity-check a proper calculation, never to size equipment.

For quick rule-of-thumb cross-checks, use our HVAC Load Calculator alongside the conversions in our HVAC Engineer's Conversion Cheat Sheet.

Frequently asked questions

Q: Manual J vs RTS — which method should I use?

Manual J for residential and small commercial under 25,000 ft². ASHRAE RTS for medium to large commercial buildings where thermal mass matters. For anything over 100,000 ft², energy code compliance typically requires EnergyPlus or DOE-2 whole-building simulation. The physics is the same — the difference is how precisely you model the time-dependent behavior.

Q: How do I handle interior spaces with no exterior walls?

Interior spaces have zero envelope load but full internal loads (people, lights, equipment) and ventilation load. In many modern deep-plan office buildings, interior zones require cooling 12 months a year because internal gains exceed envelope losses. These zones are prime candidates for heat recovery and water-source heat pump systems.

Q: Why does my calculated load not match the utility bill?

Calculated load is design-day peak. Utility bills reflect annual average conditions, which are much lower than peak. A building with a 100-ton design load might average 30-40 tons across the year. Also, actual occupancy, equipment, and operating schedules rarely match the design assumptions — one of the main drivers of the energy performance gap.

Q: How accurate do infiltration estimates need to be?

For tight commercial construction (continuous air barrier, verified by commissioning), infiltration can be as low as 0.05 ACH and may be safely ignored in a preliminary calculation. For older buildings, especially those with operable windows, infiltration can be 0.3-0.5 ACH and must be included. When in doubt, use 0.15 ACH as a conservative default for typical commercial construction.

Q: Should I include a safety factor, and if so, how much?

Energy codes including ASHRAE 90.1 generally limit oversizing to no more than 15% for cooling equipment. Apply a single 10% safety factor at the end of the calculation (after all components are summed). Do not apply separate safety factors to each individual load component — the accumulated total will be far too high.

Key takeaways

Principle	Rule
Load basis	Use ASHRAE 0.4% or 1% design conditions, not all-time records
Components	Envelope + internal (people, lights, equipment) + ventilation + infiltration
Sensible vs latent	Split the load; SHR determines coil selection and supply airflow
Safety factor	One 10-15% margin at the end, not multiple margins throughout
Oversizing	Energy codes generally limit oversizing to 15% above calculated peak
Thermal mass	RTS is 15-25% more accurate than steady-state for medium/heavy buildings
Rules of thumb	Use only for sanity checks, never for equipment selection
Ventilation	Always calculate ASHRAE 62.1 minimums — don't guess OA requirements