On this page
Size a greenhouse heater from hourly heat loss, not floor area alone. Add heat loss through the exposed roof and walls (A × U × ΔT), natural air leakage (0.02 × V × ACH × ΔT), and the base perimeter when applicable. Then compare that delivered BTU/h requirement with the heater’s rated output. Roof shape, glazing, leaks, and local winter design temperature can make two greenhouses with the same floor area need different heaters.
Use this greenhouse heat-loss worksheet
The worksheet estimates a steady hourly design load. It does not predict seasonal energy cost, warm-up time, solar gain, or the coldest temperature ever recorded. Keep every input visible so you can rerun the calculation when the covering, target temperature, or leakage condition changes.
| Term | Formula | Input you need |
|---|---|---|
| Covering and frame loss | Qconduction = Σ(A × U × ΔT) | Exposed area for each assembly, its U-factor, and temperature difference |
| Natural air leakage | Qinfiltration = 0.02 × V × ACH × ΔT | Interior volume and estimated natural air changes per hour |
| Base perimeter, when used | Qperimeter = P × F × ΔT | Ground-contact perimeter and the selected perimeter factor |
| Total delivered heat | Qtotal = Qconduction + Qinfiltration + Qperimeter | All applicable losses in BTU/h |
| Electric output equivalent | watts = Qtotal ÷ 3.412 | Total delivered BTU/h |
Calculate unlike surfaces separately. A polycarbonate end wall, inflated double-film roof, insulated knee wall, and door do not share one U-factor. Adding the four component losses is more defensible than averaging materials into one convenient number.
1. Set indoor and outdoor design temperatures
Choose the lowest indoor temperature your crop and operating plan require at night. That target is a project decision, not a universal greenhouse setpoint. For the outdoor input, use a nearby station’s heating design dry-bulb temperature rather than a normal low or a record extreme.
The ASHRAE Handbook presents 99% and 99.6% heating design conditions. In a typical year, colder conditions occur for roughly 88 hours at the 99% value and 35 hours at the 99.6% value. The more conservative choice increases heater capacity, but neither is a promise against every extreme. Match the design condition to crop consequence, backup heat, power reliability, and local professional requirements.
Subtract the outdoor design temperature from the indoor target. A 45°F target and 10°F outdoor design temperature give ΔT = 35°F. Use that same ΔT in each term unless a boundary, such as an attached heated wall, has a different temperature.
2. Measure exposed area, volume, and perimeter
For a simple gable greenhouse, record width W, length L, eave height He, and ridge height Hr. The roof slope length is not half the width; calculate the diagonal first.
| Measurement | Gable-greenhouse formula |
|---|---|
| One roof slope | s = √[(W ÷ 2)² + (Hr − He)²] |
| Both roof planes | 2 × L × s |
| Both sidewalls | 2 × L × He |
| Both end walls | 2 × [W × He + ½ × W × (Hr − He)] |
| Interior volume | W × L × He + ½ × W × (Hr − He) × L |
| Ground perimeter | 2 × (W + L) |
For a curved roof, lean-to, dome, or irregular house, use the manufacturer’s exterior surface area or calculate each real plane or arc. Do not force curved glazing into the gable formula. Floor area remains useful for layout, but it is not the heat-transfer area.
3. Choose U-factors and bracket air leakage
A U-factor describes heat transmission: a lower value means less heat loss for the same area and temperature difference. Use current documentation for the exact covering assembly first. Thickness alone does not lock the result; air spaces, cell geometry, coatings, framing, seals, and the test basis can change performance.
| Covering assembly | Fallback U-factor, Btu/(h·ft²·°F) | Use |
|---|---|---|
| Single plastic film | 1.20–1.25 | Run both ends if exact data is unavailable |
| Single glass | 1.10–1.20 | Separate insulated opaque sections |
| Separated or inflated double plastic film | 0.70–0.80 | Assumes the air layer is maintained |
| Double acrylic or polycarbonate assembly | 0.50–0.60 | Fallback only; exact panel data is stronger |
Air leakage is less certain without measurement. UGA lists 0.5–1.0 ACH for new double-layer film houses, 0.75–1.0 for new glass or fiberglass, 1.0–2.0 for older glass in good condition, and 2.0–4.0 for older glass in poor condition. A hobby polycarbonate kit does not map cleanly to those categories. Choose a defensible low and high case, calculate both, and keep the range visible.
UGA’s simplified base term uses F = 0.8 Btu/(h·ft·°F) for an uninsulated perimeter and 0.4 for its insulated condition. Treat this as a screening term, not a substitute for a site-specific ground-loss model. If a supplier or engineer provides a different method, do not add both and count the same loss twice.
Worked example: an 8 × 12 ft gable greenhouse
This is a hypothetical arithmetic example, not a measured greenhouse. Assume an 8 ft width, 12 ft length, 6.5 ft eave, 9 ft ridge, double polycarbonate fallback U = 0.60, 45°F indoor target, 10°F outdoor design temperature, 1.0 ACH, and an uninsulated base factor of 0.8.
| Step | Arithmetic | Result |
|---|---|---|
| Roof slope | √(4² + 2.5²) | 4.717 ft |
| Exposed envelope | 2(12×6.5) + 2[(8×6.5)+(½×8×2.5)] + 2(12×4.717) | 393.21 ft² |
| Interior volume | (8×12×6.5) + (½×8×2.5×12) | 744 ft³ |
| Ground perimeter | 2(8+12) | 40 ft |
| Conduction | 393.21×0.60×35 | 8,257 BTU/h |
| Infiltration | 0.02×744×1.0×35 | 521 BTU/h |
| Perimeter | 40×0.8×35 | 1,120 BTU/h |
| Total delivered heat | 8,257+521+1,120 | 9,898 BTU/h |
| Electric output equivalent | 9,898÷3.412 | 2,901 W |
The sensitivity pass changes a guess into a range. It also shows which improvement matters: lowering the target, using a better documented covering assembly, sealing known gaps, or insulating the base can be compared by changing one input and rerunning the same sheet. Do not add a generic reserve after already choosing a high leakage case unless the equipment-selection method explicitly requires it.
4. Convert the load into a heater decision
Match the result to rated delivered output. UAF Extension specifically warns to use heater output rather than input. If a combustion heater lists only fuel input, use its current certified output or exact rated thermal efficiency; do not assume an efficiency percentage. For electric resistance output, the energy conversion is 1,000 W = 3,412 BTU/h, so a 1,500 W heater corresponds to about 5,118 BTU/h.
The worked example needs roughly 2,901–3,054 W under its stated leakage cases. That means a single 1,500 W portable heater cannot meet the calculated design load. This is a design stop, not permission to improvise power. The US Consumer Product Safety Commission says portable electric space heaters should plug directly into a wall outlet, never an extension cord or power strip.
Verify the exact heater listing, nameplate, damp-location suitability, clearances, controller compatibility, receptacle, branch circuit, and locally adopted code with the manufacturer and a qualified electrician. Fixed electric space heating and portable cord-connected appliances do not share one universal “80% rule,” so this worksheet does not turn a breaker rating into an approved heater size.
Capacity alone also does not prove uniform crop-zone temperature. Route the air without blowing directly on plants, and keep the control sensor out of the heater discharge and direct sun according to exact equipment instructions. Use the temperature sensor placement guide and the broader greenhouse temperature control sequence after the load calculation.
5. Treat combustion moisture and venting as load constraints
An unvented fuel heater does more than add heat. UF/IFAS Extension notes that unvented or improperly vented gas- and oil-fired units can release carbon monoxide, ethylene, sulfur dioxide, unburned hydrocarbons, carbon dioxide, and water vapor; its guidance calls for combustion products to be vented outside and adequate combustion air to be provided.
The moisture magnitude is large enough to affect the heating decision. US DOE/PNNL guidance estimates about one gallon of liquid-equivalent water vapor per 100,000 Btu from unvented combustion. The US Energy Information Administration lists 91,452 Btu per gallon of propane. Combined, those values imply roughly 0.91 gallon of liquid-equivalent water vapor per gallon of propane burned when that moisture remains inside. This is a cross-source estimate, not a site measurement.
Do not use this article to design a flue, gas line, regulator, combustion-air opening, or unvented-heater workaround. Choose listed equipment for the actual environment and have fuel-burning systems designed and installed under current manufacturer instructions and local requirements. Plan cold-weather moisture removal with the winter greenhouse ventilation guide; ventilation adds heat demand, so the operating plan and heater design must agree.
Run this sizing audit before buying
- Lock the design temperatures. Record the indoor target, weather station, 99% or 99.6% outdoor condition, and why that risk level fits.
- Measure the real envelope. Separate roof, walls, doors, knee walls, and attached boundaries by assembly.
- Use traceable U-factors. Prefer exact current manufacturer data; bracket Extension fallbacks when it is missing.
- Calculate leakage twice. Use a low and high defensible ACH until measurement supports something narrower.
- Add perimeter loss once. State whether the simplified UGA term or another ground-loss method is used.
- Compare delivered output. Convert units, then check exact equipment output rather than fuel input or marketing area.
- Stop at the power or combustion boundary. No extension cord, power strip, DIY service upgrade, gas piping, or vent improvisation.
- Verify control and failure response. Check sensor location, heat distribution, alarm/backup plan, and seasonal readiness with the climate-control checklist.
Record both the design load and the unresolved inputs. If the answer depends on an unknown U-factor, leakage rate, electrical path, appliance output, or venting condition, the correct result is a range or an escalation—not false precision.
Sources and method
This worksheet is an editorial synthesis of UGA, UAF, UF/IFAS, ASHRAE, CPSC, EIA, and US DOE/PNNL guidance. The example was independently recalculated from its visible assumptions. My Greenhouse Lab did not measure a greenhouse or test a heater for this article; exact equipment documentation, site conditions, and qualified design remain controlling.