Modelled Versus Performance-Based Heat Load Calculations
There are three types of heat load calculations:
- Traditional rules of thumb (“1 ton per 400 sq ft”)
- Energy models (theoretical)
- Performance-based (real-world data)
Within performance-based heat load calculations, you can use energy consumption or runtime data. Energy consumption (also called energy usage or gas usage) looks at how much gas (or another fuel) is used to heat the house. Unlike rules of thumb and energy models, energy consumption is based on how the house performs under real-world conditions.
Runtime data is simply looking at how long the equipment operates at specific outdoor temperatures. If a 60,000 BTU/hr furnace runs for 30 minutes in an hour that matches outdoor design conditions, then the heating load is 30,000 BTU/hr (30 minutes / 60 minutes * 60,000 BTU/hr = 30,000 BTU/hr).
My preference is energy consumption because it’s easier to get a monthly gas bill than thermostat data. Runtime data can also be difficult to interpret for multiple-stage or variable furnaces.
Why Traditional Methods Fall Short
Traditional rules of thumb are crude guesses. They’re quick but unreliable and unlikely to provide the right answer.
Energy models aren’t much better—whether it gets you close enough depends on the accuracy of the model, the underlying assumptions, and the complete and accurate collection of household data like insulation levels, orientation, shading, air leakage, etc.
Models are conservative (they overestimate) and we often input conservative values to play it safe. That’s margin on margin.
The biggest issue with a modelled heat load is that it might be right—or wildly wrong. There’s no way to tell.
To prove my point, here’s a thought experiment: a homeowner gets an energy assessment completed. They give the report to you (the contractor) and it includes a 32,000 BTU/hr heating load. Is it an overestimate, underestimate, or close enough?
How would you know?
You could double check the report and confirm basic metrics like square footage, number of floors, location, and window count. But you won’t know the exact measurements, air leakage, insulation levels, etc. And since air leakage is the biggest source of heat loss, you can’t know if it’s accurate or not.
But if that same homeowner (located in Toronto, for my convenience) tells me they used 1,500 m³ (530 therms), I know their heating load is about 26,000 BTU/hr. Then I can recommend a 2-ton or 2.5-ton heat pump based on other factors.
Most HVAC systems are oversized because the heat loads were overestimated (with margins on margins) and the equipment has been replaced like-for-like for the life of the house. An old oversized furnace gets replaced with a new oversized furnace.
Gas Usage for Heat Loads: The Long Way
The idea is simple: a house with a furnace burns gas for heat. The more heat the house needs, the more gas it burns. So, we can look at the amount of gas used to assess how much heating the house needs.
We need four things: gas consumption, equipment efficiency, outdoor temperatures, and the 99% design temperature.
For outdoor temperatures, we’ll use a metric called heating degree days. It’s a combination of time and temperature that reflects how much heating or cooling was needed to keep an indoor temperature constant.
Outdoor temperatures are compared to a baseline temperature (usually 60°F or 65°F). If the mean temperature is 64°F for a day…well, that’s 1 degree day. While heating degree days can be counted in Celsius, we’ll need to use Fahrenheit because BTU and BTU/hr are based in Fahrenheit.
For context, Toronto has ~7,000 heating degree days with a 65°F baseline. A colder city like Edmonton has 10,000+. In US terms, think Portland, Maine (7,000 HDD) versus Anchorage, Alaska (10,000+).
Here are the steps for the heat load calculation:
- Calculate annual BTUs of heating (from m³/therms and equipment AFUE)
- Lookup heating degree days (HDD) for the time period
- Divide BTU by HDD (BTU per degree-day)
- Divide by 24 (BTU per degree-hour)
- Multiply by design/thermostat differential
- That’s your heating load!
We take the full amount of heating used (convert gas usage to millions of BTUs), taking into account equipment efficiency. Then we look up the heating degree days for our area and time period (degreedays.net is easy).
Now we divide BTU by HDD to understand how much heat (BTU) we need per degree-day. Divide again by 24 to get BTU per degree-hour.
We’re aiming for a heating load (BTU/hr), so intuitively it feels close that we have a BTU per degree-hour number. We just need to eliminate the “degree” unit—and we do that with the design temperature. Or rather, the difference between the indoor setpoint (70°F) and the design temp.
For Toronto, the 99% design temperature (found on ASHRAE) is 4°F, so the difference between indoor and outdoor temperatures will be 66°F (70 minus 4 equals 66).
If our Toronto house needed 360 BTU per degree-hour, then the heating load is ~24,000 BTU/hr (360 * 66 = 23,760).
That’s the long way of doing it (although significantly faster than energy modelling). Tools like thermalpoint.ca, knowyourload.ca, and thermentor.com are making it easier and faster.
How This Affects Your Heat Pump Sizing
Getting the heat load right is critical for properly sizing heat pumps. As Gary notes in his heat pump installation guide, ductwork constraints often limit how large your heat pump can be. If you size strictly to an overestimated heat load, you may encounter airflow problems.
“If a home has a heat loss of 60k BTU and a heat gain of 24k BTU, how do we size? A heat pump will need 400-450 CFM per ton to run effectively. If we size to the heating load, we need 2000-2250 CFM. In most retrofit applications, we’ll find ductwork only designed to carry 800-1200 CFM.”
The solution is to size closer to the cooling load but as close to the heating load as possible, then supplement with auxiliary heat as needed. This is exactly why accurate heat load calculations are so important.
Simplifying the Process
For contractors and homeowners who want to skip the manual calculations, several online tools make this process much simpler. But the principle remains the same: using actual energy consumption data will generally give you a more accurate heat load estimate than theoretical models alone.
Accurate heat loads lead to properly sized heat pumps, which avoid the comfort issues, short cycling, and poor dehumidification that come with oversized equipment.
This article is Part 2 of a 3-part series on heat load calculations and proper HVAC sizing by Drew Towzer for HVAC Know It All.