By Erik Jannsen
Heat pumps are changing the landscape, and heat loss calculations are becoming a critical part of the conversation in retrofits. In most existing homes, furnaces are sized based on what’s already there; if the old unit worked and wasn’t clearly too oversized, a similarly sized unit goes in, but that approach doesn’t work for heat pumps.
Replacing a 60,000 BTU/h furnace with a comparable five-ton heat pump will likely lead to a cascade of issues beyond unnecessarily high upfront costs: short cycling, poor temperature control and dehumidification, premature component wear, and faults or noise related to an airflow mismatch with the ductwork.
To get the sizing right, we need to go beyond rules of thumb, considering both proven heat loss calculation methods from new construction and newer approaches based on real-world data from the home.
The established method for calculating residential heat loss in Canada is based on CSA F280-12. It uses home dimensions, component specs, insulation values, air infiltration, and other inputs to calculate the design heat loss (DHL) at the outdoor design temperature (ODT) for heating. To be clear, CSA F280-12 doesn’t consider the energy use of the home. It’s also conservative and this ensures the home won’t be cold.
In new construction, CSA F280-12 is more straightforward; drawings and specifications are readily available. In retrofits, it’s more complex. You need to gather the needed information and explore what’s behind the walls or make educated guesses. Air infiltration, a key input, requires a blower door test to estimate accurately. EnerGuide audits help with this information, but many homeowners skip them without incentives.
The quoting process adds another layer. Contractors provide free quotes, but accurate heat pump sizing requires a heat loss estimate, a sunk cost with no guarantee of winning the job. Solutions are needed.
Emerging approaches
The path to faster retrofit heat loss calculations splits in at least two directions: speeding up CSA F280-12, and exploring alternative methods based on real-world data.
Several innovators are simplifying CSA F280-12 compliance. Volta SNAP, developed by Toronto-based Volta Research, uses EnerGuide audit results to generate a compliant calculation. Properate, based in B.C., has developed an F280-certified app called Stride that scans interior rooms and builds a 3D model with all needed dimensions — saving significant time.
Additionally, it’s possible to leverage real-world data from the home. Empirical means based on data, and two promising sources for empirical heat loss methods are smart thermostat data and energy bills.
These methods are intuitive. If a home has a single-stage 60,000 BTU/h furnace and thermostat data shows it only ran 50 per cent of the time overnight near the outdoor design temperature (ODT), the actual heat loss is likely closer to 30,000 BTU/h.
Energy bills also help. Heat loss and energy use are clearly related — higher loss means higher consumption. The relationship just needs untangling.
One approach was highlighted by the Green Building Advisor and showcases a 15-minute approach to determine the building’s design heating load. From there, using a calculator, like knowyourload.ca, it can implement it for Canadian homes.
Another tool, ThermalPoint, developed by the Toronto and Region Conservation Authority with support from collaborators, The Atmospheric Fund and Voltage Research, takes a similar approach. Currently, for gas-heated homes in Ontario (with plans to expand), it uses annual gas consumption and a few other inputs to estimate both heat loss and key heat pump sizing metrics.
ThermalPoint reverses traditional energy modelling. Instead of using heat loss and weather data to predict energy use, it uses gas consumption and weather data to estimate heat loss. The result reflects average home heat loss at different outdoor temperatures. In contrast, CSA F280-12 focuses on the extremes.
The above image shows ThermalPoint inputs for an empirical heat loss calculation. The annual gas consumption is the key input, and it is available from the homeowner utility portal.
ThermalPoint estimates a heating load line (red) for the home, and overlays on top of it an optional user-defined heat pump capacity curve (blue). The point where they intersect is a threshold below which backup or supplemental heating will likely be needed (-9 C in this example).
Empirical heat loss in action
To test both ThermalPoint and CSA F280-12, Keith Burrows, director of low-carbon buildings at the Atmospheric Fund, used both testing approaches when sizing a heat pump for his home. “The ThermalPoint algorithm gave me and my installer the confidence to size the system much more aggressively than the CSA F280 calculations suggested. The CSA F280 calculations are conservative, and the heat loss results for my house were 65 per cent
higher than ThermalPoint. Ultimately, we sized the heat pump using the ThermalPoint results as a guide, with enough supplemental electric resistance to meet the CSA F280 calculations. The result is a heat pump that hit the sweet spot: it never short-cycles and only requires supplemental electric resistance a few times a year when outdoor temperatures get really cold,” shared Burrows.
Another example of in-field use for these methods was at Victor Hyman’s, executive director of ClimateCare, home. “I was surprised when ThermalPoint produced an empirical load calculation of half of the EnerGuide Report load done on my house,” explained Hyman. “The ThermalPoint calculation correlated with the design heating day runtime data from my Ecobee, and I chose to size my hybrid system based on ThermalPoint. I removed a two-stage 80,000 BTU/h furnace and installed a two-stage 40,000 BTU/h furnace, along with a two-ton cold climate air-sourced heat pump. After this winter, I can attest that it nailed the sizing. My system ran as ThermalPoint predicted, and my house has never been so comfortable. I continue to use and recommend empirical load calculations as a check in retrofit applications.”
Comparing methods
CSA F280-12 is standardized, defensible, and sometimes required in retrofits for code or incentives. It’s the only method suitable for room-by-room heat loss, which is essential for ductless systems, and helps evaluate energy upgrade impacts. But it takes time and skill, and in retrofits, accuracy depends on the quality of the information that was gathered.
Empirical methods, by contrast, can take a few minutes. They use simplified models, but they are calibrated using real data. While they can’t account for other energy upgrades, they reflect how the home performs today. However, they rely on specific data, which may not always be available. Certain home-specific factors can also skew results, so contractors must judge suitability case by case.
A follow-up article will evaluate different heat loss methods applied in retrofits. Early results suggest empirical methods tend to better predict the outdoor temperature at which backup or supplemental heat is needed, as well as overall heat pump utilization.
These methods aren’t competing; they can be complementary. Empirical approaches can help properly size the heat pump to handle most of the load, while CSA F280-12 remains important for sizing supplemental heating systems to maintain comfort during design conditions.