Greenhouse Heating Load Calculator

Delivered heating capacity

Heat loss {{ resultsReady ? fmt(result.total) + ' kW' : '—' }} Planning estimate

Roof shape:

Gable Hoop Single slope

Semicircular, ground-to-ground hoop; peak height is half the width.

Freestanding structure with all four walls exposed.

Length:

Width:

Eave height:

Height of the vertical wall, below the roof.

Roof rise:

Vertical rise above the eaves; not the total ridge height.

Covering:

Covering U-value:

Use the overall assembly value, including surface effects.

Inside target:

Outside design temperature:

Use your site’s winter design temperature, not the current forecast.

Air leakage:{{ params.ach }}

Estimate winter leakage with doors and vents closed.

ACH

Ground perimeter:

Choose the ground-edge insulation condition. Exclude edge loss only when that boundary is handled separately.

Screen reduction:{{ params.screen }}

Reduction of covering loss only. Enter evidence for your installed screen; 0 means none.

Capacity allowance:{{ params.margin }}

Extra delivered heater capacity above the design heat loss.

Sensitivity range:{{ params.uncertainty }}

Symmetric planning variation around heat loss; not a statistical confidence interval.

Sensitivity range unavailable. Enter 0–100%; heating load and heater capacity remain available.

Candidate delivered output:

Optional heater comparison; 0 means no candidate.

Check the optional heater output; heating load remains available. Enter 0 to skip.

Scenario fuel:

Choose a fuel unit, then enter its matching price and heater efficiency.

Heater efficiency:

Delivered heat divided by fuel input. Electric resistance is approximately 100%; enter combustion efficiency.

Energy price:

Enter your price for the selected fuel unit; this example is not a market quote.

USD/{{ fuelInfo.unit }}

Equivalent full-load hours:

Hours at the design heat loss, not thermostat or burner runtime.

h/day

Scenario duration:

Number of days in this entered energy scenario.

days

Metric	Value	Unit	Basis	Copy
{{ row[1] }}	{{ typeof row[2] === 'number' ? fmt(row[2]) : row[2] }}	{{ row[3] }}	{{ row[4] }}

Chart unavailable. The heat-loss ledger still contains the values.

Metric	Value	Unit	Basis	Copy
{{ row[1] }}	{{ typeof row[2] === 'number' ? fmt(row[2]) : row[2] }}	{{ row[3] }}	{{ row[4] }}
Energy scenario unavailable: {{ result.energy_message }} Heating load remains available.

A greenhouse can be warm in afternoon sunshine and still need continuous heating on a cold night. Once the desired indoor temperature has been reached, the heater must replace heat as it escapes. The rate needed to maintain that temperature under a chosen cold-weather condition is the design heating load. It provides a starting point for comparing heaters, improving insulation, and planning a heating-energy budget.

Floor area alone does not determine that load. Heat passes through the exposed walls and roof, leaves with leaking warm air, and escapes around the ground edge. A taller greenhouse can have more covering and more enclosed air than a lower structure with the same footprint. The covering's overall U-value describes how readily heat passes through each unit of area. A lower U-value means less heat loss at the same temperature difference.

Heat loss: The rate at which heat must be replaced, expressed in kilowatts (kW) or British thermal units per hour (Btu/h).
Delivered heating capacity: The useful heat a heater can supply to the greenhouse. This is different from the energy rate entering a combustion heater as fuel.
Energy use: The accumulated energy supplied over time, often expressed in kilowatt-hours (kWh). A heater's capacity alone does not establish how much energy it will consume over a season.

Improving the covering, sealing leaks, and insulating the ground edge address different losses. Air leakage is commonly described in air changes per hour (ACH), meaning the volume of replacement air entering each hour relative to the greenhouse's enclosed volume. Better glazing does not automatically make doors or vents airtight. Likewise, a thermal screen's reduction in covering loss should not be read as the same percentage reduction in the whole heating bill.

The indoor temperature should reflect the plants being grown, while the outdoor design temperature should represent the cold conditions the heating system is intended to handle. A larger difference increases the heat required. Wind and the condition of joints and closures can change leakage, so measurements and construction details matter as much as the nominal covering material.

A steady-state estimate holds these conditions constant. It does not predict how quickly a cold greenhouse will warm up or how stored heat and sunshine will change demand through the day. Equipment capacity and seasonal energy therefore need different assumptions. Neither a heat-loss total nor an added capacity allowance establishes uniform temperatures throughout the crop or protection during an equipment failure.

How to Use This Tool:

Set up the structure and winter temperatures first. Results update as you edit, and Advanced adds screen adjustments, heater comparisons, and an optional energy scenario.

Choose Roof shape and enter the dimensions. Gable assumes a symmetric roof. Hoop assumes a ground-to-ground semicircle with a peak height equal to half the width. Single slope assumes a freestanding building with all four walls exposed. For the two straight-roof shapes, Roof rise is the vertical increase above the eaves, not the total height. For a single slope, Eave height is the lower wall height.
Select Covering, estimate Air leakage with doors and vents closed, and choose the Ground perimeter condition. Choose Custom U-value when you have an overall assembly value for your covering. Use Exclude edge loss only when that boundary is accounted for separately.
Enter Inside target and Outside design temperature. Supply the site's winter design temperature rather than its current forecast. Each dimensional and temperature input has its own unit choice, so mixed units are supported. Changing a unit converts the entered measurement rather than merely changing its label.
Open Advanced to enter an evidence-based Screen reduction, an optional Capacity allowance, and a Sensitivity range. For a heater comparison, enter its Candidate delivered output in kW or Btu/h. Enter 0 to skip the comparison rather than leaving that field blank.
Choose Scenario fuel and enter the matching efficiency, price, equivalent full-load hours per day, and duration for an energy estimate. Prices are in US dollars (USD) per selected fuel unit. Changing the fuel does not automatically replace your efficiency or price, so review both values.
Check Design heat loss and Delivered heater output in the Heat-loss ledger before comparing equipment. If the headline says Check inputs, correct the value named in its message. Required-input errors withhold the heating result; errors in optional sensitivity, heater comparison, or energy inputs leave the primary heat-loss calculation available.

Interpreting Results:

Design heat loss is the estimated ongoing demand at the entered temperatures. The headline Delivered heating capacity and the ledger's Delivered heater output include the chosen capacity allowance. Compare these with a manufacturer's delivered-output rating at the relevant operating conditions, not just a fuel-input nameplate.

A positive Candidate capacity difference means spare output above the allowance-adjusted estimate; a negative value means a shortfall. This comparison does not use the sensitivity upper bound. A heater can meet the selected capacity estimate while remaining below that upper bound. The sensitivity range is an entered variation around heat loss, not a statistical confidence interval.

Loss breakdown shows whether covering, air leakage, or perimeter loss contributes most under the current assumptions. The Screen reduction result is covering heat saved at the design temperature difference, not a measured seasonal saving. Verify the covering and leakage assumptions before treating the largest contribution as a spending priority.

Heater input rating includes the capacity allowance and divides delivered capacity by efficiency. Peak operating fuel, daily energy, and scenario costs instead use design heat loss without that allowance. They are not calculated from the candidate heater's size. A larger allowance or candidate therefore does not, by itself, increase the energy scenario.

Zero heat loss means the entered outside temperature is at or above the inside target. It does not establish that heating is unnecessary at other times, and it is not a cooling-load or ventilation calculation.

Technical Details:

Steady-state heat balance adds covering loss, air-leakage loss, and a simplified ground-edge loss. All three use the same positive indoor-to-outdoor temperature difference. One overall covering U-value applies to the exposed walls and roof, including the end walls.

Covering, air leakage and ground-edge heat loss

The temperature difference is limited to zero or greater, and the three heat-loss terms are added in kW.

\begin{array}{l} ΔT = \max (0, T_{i} - T_{o}) \\ Q_{c} = \frac{U A ΔT (1 - s)}{1000} \\ Q_{a} = \frac{a V n ΔT}{1000} \\ Q_{e} = \frac{b P ΔT}{1000} \\ Q = Q_{c} + Q_{a} + Q_{e} \end{array}

Quantities used in the greenhouse heat-loss equations
Symbol	Meaning	Unit
T_i, T_o	Inside target and outside design temperature	°C
ΔT	Positive temperature difference	K
A	Exposed walls and roof, excluding the floor	m²
V	Enclosed air volume	m³
P	Ground perimeter length	m
U	Overall covering heat-transfer coefficient	W/(m²·K)
n	Air leakage in air changes per hour	h⁻¹
s	Entered screen-reduction percentage divided by 100	Fraction
Q_c, Q_a, Q_e, Q	Covering, leakage, perimeter, and total design heat loss	kW

The coefficients a and b are defined below. Dividing each watt result by 1000 gives kW. Screen savings equal the unscreened covering loss minus the reduced covering loss. The screen percentage does not change enclosed volume, air leakage, or perimeter loss.

Geometry of the exposed enclosure

Let L be length, W width, H eave height, and r roof rise, all in metres. The floor is not part of exposed covering area. For every supported shape, the ground perimeter is the rectangle around the footprint.

P = 2 (L + W)

Symmetric gable

Two sloping roof faces sit above vertical walls. The two triangular end sections together contribute W times r to covering area.

\begin{array}{l} A = 2 L H + 2 W H + W r + 2 L \sqrt{{(\frac{W}{2})}^{2} + r^{2}} \\ V = L W (H + \frac{r}{2}) \end{array}

Ground-to-ground semicircular hoop

The curved covering is half a cylinder, and the two semicircular ends together form one circle. Eave height and roof rise are not used.

\begin{array}{l} A = \frac{π W L}{2} + \frac{π W^{2}}{4} \\ V = \frac{π W^{2} L}{8} \end{array}

Freestanding single slope

The long walls have heights H and H plus r. Both end walls and the full sloping roof remain exposed; no shared wall is subtracted.

\begin{array}{l} A = L (2 H + r) + W (2 H + r) + L \sqrt{W^{2} + r^{2}} \\ V = L W (H + \frac{r}{2}) \end{array}

Covering resistance and loss coefficients

The covering and leakage assumptions follow University of Georgia Cooperative Extension Bulletin 792, a U.S. greenhouse estimate revised November 4, 2025. The covering presets use overall assembly R-values, which describe resistance to heat flow and include surface effects. Their reciprocal gives an imperial U-value. With the rounded power conversion used here, the U-value in W/(m²·K) is calculated as follows, where R_t is in h·ft²·°F/Btu.

U = \frac{1.8}{{0.3048}^{2} \times 3.412 \times R_{t}} \approx \frac{5.6784990475}{R_{t}}

Covering preset resistance values and their approximate metric U-values
Covering	R_t, h·ft²·°F/Btu	U, W/(m²·K)
Single glass	0.91	6.2401
Double glass, ¼-inch air gap	2.00	2.8392
Single plastic film	0.83	6.8416
Separated double plastic film	1.43	3.9710
Double film over glass	2.00	2.8392
Fiberglass panel	0.83	6.8416
Double acrylic / polycarbonate	2.00	2.8392

These are approximate covering assumptions, not product-specific ratings. A custom U-value replaces the preset. If that value already includes the same thermal-screen benefit, applying an additional screen reduction would count the benefit twice.

The leakage coefficient converts the fixed imperial factor of 0.02 Btu/(ft³·°F). The perimeter coefficient converts an imperial edge factor f in Btu/(h·ft·°F), using 0.8 for an uninsulated edge, 0.4 for an insulated edge, or 0 when excluded.

\begin{array}{l} a = \frac{0.02 \times 1.8}{{0.3048}^{3} \times 3.412} \approx 0.3726049244 \\ b = \frac{f \times 1.8}{0.3048 \times 3.412} \end{array}

The units of a are Wh/(m³·K). The resulting b values are approximately 1.3846452077 W/(m·K) for an uninsulated edge and 0.6923226039 W/(m·K) for an insulated edge. Selecting an insulated edge therefore halves only the perimeter term. This term uses the outdoor-air temperature difference, not a separately entered soil temperature.

Indicative leakage estimates are about 0.5 to 1 ACH for new double-film construction and 2 to 4 ACH for old glass in poor condition. Wind and maintenance affect those estimates. Covering selection does not automatically set ACH, and no separate wind-speed adjustment is calculated.

Capacity allowance, sensitivity, and heater comparison

Let m be the capacity-allowance percentage divided by 100 and z the sensitivity percentage divided by 100. Estimated delivered capacity C is based on design heat loss Q, while sensitivity bounds vary Q independently.

\begin{array}{l} C = Q (1 + m) \\ Q_{low} = Q (1 - z) \\ Q_{high} = Q (1 + z) \\ D = C_{h} - C \\ ε = 10^{- 10} \times \max (1, C, C_{h}) \end{array}

C_h is candidate delivered output, and D is its unrounded difference from C. All capacity values and the numerical tolerance ε are in kW. A candidate of 0 skips comparison. For a valid positive candidate, differences within the tolerance are treated as zero.

Candidate delivered-output comparison boundaries
Unrounded difference	Comparison	Reported difference
D < −ε	Below estimated capacity	Negative shortfall
−ε ≤ D ≤ ε	Meets estimated capacity	Zero
D > ε	Meets estimated capacity	Positive spare output

A 0% sensitivity setting makes both bounds equal to Q; 100% produces bounds from zero to twice Q. Neither setting establishes the likelihood of those outcomes. The allowance and sensitivity range do not alter the underlying heat-loss terms.

Fuel input and assumed energy use

Efficiency η is delivered heat divided by fuel-input energy, entered as a percentage and divided by 100. Let h be equivalent full-load hours per day, d scenario days, e fuel heat content in kWh per priced unit, and p price in USD per unit. Heater input rating I corresponds to delivered capacity C; operating fuel and energy instead follow design heat loss Q.

\begin{array}{l} I = \frac{C}{η} \\ F_{peak} = \frac{Q}{η e} \\ E_{day} = \frac{Q h}{η}, E_{scenario} = E_{day} d \\ F_{day} = \frac{E_{day}}{e}, F_{scenario} = \frac{E_{scenario}}{e} \\ {Cost}_{day} = p F_{day}, {Cost}_{scenario} = p F_{scenario} \end{array}

I is in kW, F_peak is in fuel units per hour, E denotes input energy in kWh, F denotes fuel quantity, and cost is in USD. The following fixed sample heating values determine e by dividing Btu per unit by 3412.

Fuel heating values used for the energy scenario
Fuel	Priced unit	Btu per unit	Approximate kWh per unit
Electricity	kWh	3,412	1
Natural gas	therm	100,000	29.308324
Propane	US gallon	91,452	26.803048
Heating oil	US gallon	138,500	40.592028

The displayed kWh equivalents above are rounded; the calculation divides the listed Btu values by 3412 without that intermediate rounding. Actual fuel heat content can vary. Combustion efficiency should use the matching higher-heating-value (HHV), or gross heat-content, basis. Electric resistance is represented by an efficiency near 100%; heat-pump performance is not modeled.

Equivalent full-load hours represent energy at the design heat loss, not thermostat-on time or burner runtime. The same entered daily equivalent repeats for every scenario day. Zero hours makes daily and scenario energy zero; zero days makes only the scenario totals zero; zero price makes costs zero without removing energy demand. Fixed charges are excluded.

Units, input boundaries, and precision

Length conversions use 0.3048 metres per foot. Fahrenheit temperatures include the 32-degree offset, but a temperature difference does not. The power conversion is the rounded value of 3412 Btu/h per kW, not a more precise conversion substituted from another source.

\begin{array}{l} ℓ_{m} = 0.3048 ℓ_{ft} \\ T_{°C} = \frac{T_{°F} - 32}{1.8} \\ {ΔT}_{K} = \frac{{ΔT}_{°F}}{1.8} \\ C_{Btu/h} = 3412 C_{kW} \end{array}

One kelvin of temperature difference equals one Celsius degree of difference. A custom U-value in Btu/(h·ft²·°F) is multiplied by approximately 5.6784990475 using the same conversion basis. Candidate Btu/h is divided by 3412 before comparison.

The following limits are inclusive and apply after unit conversion. They are accepted input ranges, not evidence that every building or operating condition within them is suitable for this simplified model.

Accepted numeric input ranges after conversion to the listed units
Input	Minimum	Maximum	Unit or condition
Length and width, each	0.1	1000	m
Eave height	0.1	30	m; gable and single slope
Roof rise	0	30	m; gable and single slope
Custom covering U-value	0.05	20	W/(m²·K)
Inside and outside temperature, each	−80	80	°C; equivalent to −112 to 176 °F
Air leakage	0	20	ACH
Screen reduction	0	80	%
Capacity allowance and sensitivity, each	0	100	%
Candidate delivered output	0	100,000	kW; maximum 341,200,000 Btu/h
Heater efficiency	1	100	%
Energy price	0	1000	USD per selected fuel unit
Equivalent full-load hours	0	24	h/day
Scenario duration	0	366	days

Numeric entries may contain decimals. A blank or invalid sensitivity value makes its bounds unavailable rather than assuming 0%. A blank or invalid candidate removes its comparison; entering 0 deliberately skips it. Invalid fuel, efficiency, price, hours, or duration makes the entire energy scenario unavailable, including its input-rating result. These optional errors do not remove valid design heat loss or delivered capacity.

Calculations retain full precision until display. Most numeric results show two decimal places, while the headline Btu/h equivalent normally shows whole numbers. Nonzero values smaller than the display increment use scientific notation with up to three significant digits. Candidate comparison uses unrounded capacities and the stated tolerance, not equality of the displayed rounded numbers.

Limitations:

This estimate supports preliminary comparisons under declared conditions. Its main omissions are:

Changing heat demand from solar gain, thermal mass, warm-up, crop evaporation, humidity control, and a door-opening schedule. No weather history or forecast determines the entered full-load hours.
Attached lean-tos, multi-span buildings, non-semicircular hoops, and separate wall or roof assemblies. Ground loss is only a perimeter approximation, not a buried-floor conduction calculation.
Heating distribution, thermostat behavior, installation suitability, and backup capacity. Energy costs use a constant entered price and efficiency rather than predicting fuel prices, cycling losses, or a utility bill.

Final heater, electrical, combustion, ventilation, and backup design need local professional review. Confirm site design temperature, covering data, leakage, and manufacturer delivered output before selecting equipment. A numerical capacity match or user-chosen allowance is not an engineering safety guarantee.

Worked Examples:

A small gable greenhouse and a 6 kW heater

For a 6 m by 3 m gable greenhouse with 2 m eaves and a 1 m roof rise, exposed area is approximately 60.6333 m², enclosed volume is 45 m³, and perimeter is 18 m. Separated double plastic film gives a U-value of approximately 3.970978 W/(m²·K). With 15 °C inside, −5 °C outside, 1 ACH, an uninsulated edge, and no screen reduction, covering, leakage, and perimeter losses are approximately 4.815471, 0.335344, and 0.498472 kW. Their sum displays as Design heat loss of 5.65 kW. A 10% allowance raises Delivered heater output to 6.21 kW, so a 6 kW candidate falls short by about 0.21 kW. Entering 20% sensitivity shows 4.52 to 6.78 kW around heat loss; it does not raise the selected 6.21 kW capacity to that upper bound.

A screened greenhouse with a propane scenario

A gable greenhouse 20 ft long and 8 ft wide, with 6 ft eaves and a 3 ft rise, has 560 ft² of exposed covering, 1200 ft³ of air volume, and a 56 ft perimeter. Enter a custom U-value of 1 Btu/(h·ft²·°F), 60 °F inside, 30 °F outside, 2 ACH, and an uninsulated edge. A 25% screen reduction lowers covering loss from 16,800 to 12,600 Btu/h. Leakage adds 1440 Btu/h and the edge adds 1344 Btu/h, giving 15,384 Btu/h, or about 4.51 kW, of design heat loss. A 10% allowance gives 4.96 kW delivered capacity. At 80% efficiency, the corresponding input rating is 6.20 kW. For 8 equivalent full-load hours per day over 120 days, propane use is approximately 201.86 US gallons. An illustrative price of USD 2.50 per US gallon gives a Scenario energy cost of USD 504.66. That fuel total uses the 4.51 kW heat loss, not the larger delivered-capacity allowance.

References:

Greenhouses: Heating, Ventilation, and Cooling, Bulletin 792, University of Georgia Cooperative Extension, November 4, 2025.
British thermal units (Btu), U.S. Energy Information Administration.
Glossary: Heat content, U.S. Energy Information Administration.