Internal heat gains are the heat produced inside a building by the people who use it, the lights that illuminate it and the equipment plugged into it. In many commercial buildings, especially interior zones, offices with modern glazing and anything with dense occupancy, internal gains are the largest cooling load.
They are also the inputs most often guessed. Default values carried forward from older projects frequently overstate lighting and equipment loads, and occupancy densities are often taken from code maximums rather than actual use. This guide explains each source, gives typical ranges from established references, and shows how to apply them sensibly.
People
Occupants release both sensible heat (by convection and radiation) and latent heat (by breathing and perspiration). The amount depends on activity level. ASHRAE Handbook, Fundamentals tabulates heat gain by activity. The values below are representative of that table for a room at typical comfort conditions; the exact split changes with room temperature.
| Activity | Approximate sensible gain per person | Approximate latent gain per person |
|---|---|---|
| Seated, at rest (theatre) | about 60 to 70 W | about 30 to 40 W |
| Seated, light office work | about 70 to 75 W | about 45 to 60 W |
| Standing, light work (retail) | about 75 to 90 W | about 55 to 75 W |
| Moderate work (light bench work) | about 80 to 100 W | about 100 to 140 W |
| Heavy work or exercise | 140 W or more | 250 W or more |
Occupant density
Density matters as much as heat per person. A code occupant load used for exit design is a maximum, not a typical value. An open office might be designed around one person per 10 to 15 m², while a meeting room or classroom can be much denser. Use the owner's planned occupancy where known, and the default densities in ASHRAE Standard 62.1 or NECB modelling rules as a fallback.
Time lag
Roughly half of the sensible gain from people is radiant, which is absorbed by surfaces and released later. Latent gain becomes a load immediately. This is handled by the cooling load method; see commercial cooling load calculations.
Lighting
All electrical power to lighting eventually becomes heat in the building. The gain is:
Lighting heat gain (W) = installed lighting power (W) x usage factor x special allowance factor
The usage factor reflects how much of the installed lighting is on at the design hour. The special allowance factor accounts for ballast or driver losses, which are small for modern LED fixtures.
Lighting power density
Lighting loads have fallen dramatically with the move to LED and the tightening of energy codes. NECB 2017 and 2020 set maximum lighting power allowances by space or building type, and modern LED designs commonly come in below those limits. Older fluorescent installations often had lighting power densities double or more of current designs. When retrofitting, measure or count fixtures rather than relying on the original design values.
Where the heat goes
With recessed fixtures and a return air plenum, part of the lighting heat goes directly into the plenum and is picked up by return air, reducing the room load but not the coil load. With pendant or surface fixtures, nearly all heat enters the room. ASHRAE publishes space fraction and radiant fraction values for common luminaire types.
Equipment and plug loads
Equipment gains include computers, monitors, printers, kitchen appliances, laboratory equipment, motors and process loads.
Office equipment
Office plug loads have generally decreased as desktop computers have given way to laptops and LCD monitors. ASHRAE research distinguishes light, medium and heavy load density offices. Typical design values for general offices are often in the range of 5 to 15 W/m² (about 0.5 to 1.4 W/ft²), but this should be confirmed against the actual tenant use. Nameplate ratings greatly overstate actual consumption and should not be added up directly.
Kitchens and labs
Commercial kitchen equipment produces large sensible and latent loads. ASHRAE provides values for hooded and unhooded appliances; under a properly functioning hood, most convective and latent heat is exhausted and only the radiant portion reaches the room. Laboratory equipment varies widely, and fume hoods themselves are exhaust devices, not heat sources. Get real equipment lists for these spaces.
Motors
Motors located in the conditioned space add their full input power. Motors outside the space driving equipment inside (or vice versa) add only part. This is relevant for fan heat in air handlers and pumps in mechanical rooms.
IT equipment
Server rooms and data halls are dominated by IT load, which is almost entirely sensible. Design is typically based on the planned IT power, not nameplate. See server room and telecom closet cooling.
Schedules and diversity
Internal gains rarely peak together. An office is seldom at full occupancy; many people are in meetings, travelling or working remotely. Lights may be dimmed by daylight controls. Equipment cycles. Load calculations should use:
- Hourly schedules for occupancy, lighting and equipment.
- A realistic peak fraction for each schedule, not 100% by default.
- Separate diversity for zone sizing (where local peaks matter) and central plant sizing (where building-wide diversity applies).
Our guide on diversity and safety factors in load calculations covers how to apply these without double counting.
Common mistakes
- Using nameplate equipment ratings instead of measured or published operating values.
- Applying legacy lighting densities to an LED design.
- Using maximum code occupant loads for every room at the same time.
- Ignoring internal gains in heating calculations when they are reliable and continuous, or crediting them when they are not.
- Forgetting heat from in-space motors or process equipment.
- Applying a safety factor to internal gains and again to the total load.
Internal gains and energy
Internal gains reduce heating needs in winter and increase cooling needs in summer. In large, well-insulated buildings, interior zones can need cooling year-round. That makes internal gains central to energy modelling and to opportunities like heat recovery chillers that move heat from the core to the perimeter.
Frequently asked questions
Should I use nameplate power for computers?
No. Nameplate ratings are maximums for electrical safety and are typically far higher than actual average consumption. Use published measured values or metered data.
Do daylighting controls reduce cooling loads?
Yes, when lights are actually dimmed during peak cooling hours. The effect should be modelled with schedules rather than assumed as a flat reduction.
How do I estimate loads for a building with no tenant yet?
Use default values from ASHRAE or NECB modelling rules for the expected space type, and document the assumption. Provide flexibility in zoning and terminal units rather than inflating central plant capacity.
Key takeaways
- People produce both sensible and latent heat that varies with activity.
- Lighting loads are much lower with LED; verify actual installed power.
- Plug loads should be based on measured or published values, not nameplate.
- Schedules and realistic peak fractions prevent oversizing.
- Accurate internal gains improve both load calculations and energy models.
Talk to an engineer
Have a project or a problem this guide touches on? Tell us about the building and we’ll be in touch within one hour.