Manual J vs. the New Wave: Why Classic Load Calcs Still Reign Supreme

Thermentor and runtime analytics have promise, but Manual J remains the only fully defensible method for HVAC design—especially in high-performance homes.

The Heat Pump Boom and the Sizing Dilemma
The electrification wave is here, and heat pumps are riding the front. As more homes ditch fossil fuels for efficient, all-electric systems, accurate HVAC sizing has never been more critical. Oversize it, and you risk poor dehumidification, short cycling, and higher bills. Undersize it, and comfort goes out the window.

In the middle of this shift, a new breed of load estimation tools is emerging. Platforms like Thermentor and other runtime-data-driven methods offer a shortcut: why measure or model anything when you can just "read the energy bill?" It's tempting. But here’s the truth: for all their convenience, these tools still offer an incomplete picture compared to the precision and physics-based rigor of a Manual J calculation.

Let’s unpack why.

Manual J: The Foundation of HVAC Design
Manual J, developed by ACCA, is the industry standard for a reason. It factors in every component that contributes to heating and cooling load:

• Geographic design conditions (99% and 1% temperatures)

• Orientation and solar gains

• Wall, floor, roof, and window assemblies (with verified R/U-values)

• Air infiltration (ideally measured via blower door)

• Internal gains and occupancy assumptions

• Mechanical ventilation requirements

The outcome is a room-by-room, hour-by-hour peak design load—not an average, not a single-season estimate, and definitely not a heuristic.

It used to be slow and tedious in an existing building. Modern apps like Conduit Tech have changed that . And it’s the most complete and defensible way to do it.

And with the advent of variable-speed heat pumps, this level of detail is more important than ever. We're not just designing for the coldest sensible day anymore—Manual J also includes peak latent loads for the 1% cooling design day. That means we aren’t only looking at temperature, but also how well our system can handle humidity when it really matters.

This is often misunderstood or overlooked by runtime-focused approaches. While those methods may claim precision based on actual consumption, they rarely capture the full moisture dynamics or the psychrometrics needed to predict latent performance. Unless your runtime data includes psychrometric tracking and a robust latent performance curve, you're flying blind on moisture control.

The Case for Thermentor: A New Kind of Estimation
Thermentor, powered by Clean Power Research’s Virtual Energy Audit, analyzes monthly energy bills and weather data to infer a building’s thermal performance. Its biggest strengths:

• Reflects real-world energy usage, not assumptions

• Low labor input—no blower door or onsite audit required

• Gives a snapshot of system performance across an actual season

This can be especially useful for retrofit scenarios where data from an existing system is available. It may provide a more precise view of seasonal sensible loads than a Manual J—but that precision is inherently limited in scope. It is precise within an incomplete frame of reference: it does not account for edge-case design days, latent demand, or untested operational modes.

But Here's the Catch: Partial Data Isn’t Design Data
The Thermentor white papers themselves acknowledge a few key limitations:

1. Historical Data May Not Reflect Design Conditions

   • If past winters or summers were mild, the loads will be understated. Design sizing must account for extremes.

2. Data Granularity Is Limited

   • Monthly utility bills mix HVAC, lighting, hot water, and plug loads. Without submetering, attribution to HVAC is inferred.

3. No Latent Load

   • Cooling isn’t just about temperature. Humidity is load too. Runtime data doesn’t show whether a system ever achieved proper dehumidification.

4. No Use in New Construction

   • No past bills? No analysis. Runtime-based methods can’t be used in most new builds or aggressive retrofits.

Manual J Is Conservative – That’s by Design
Manual J uses standardized inputs for solar and internal gains, and applies design-day conditions to ensure equipment is prepared for peak sensible and latent demand. It’s not guessing what happened last season—it’s planning for what could happen on the most demanding day.

And it's not just about the coldest day in winter. Manual J also ensures that your system can handle the stickiest, sweatiest summer afternoons with proper latent removal—those 1% humidity-loaded days when a shortcut leads to mold, condensation, and callback city.

In Thermentor's comparison of 17 homes, Manual J heating loads appeared higher than the observed loads calculated from utility data. However, this data is not without caveats. The study derived 'actual' load from electric resistance system runtime and 15-minute interval utility usage on the coldest day of the year. No verification was done to confirm if the homes were fully conditioned, if all rooms were actively heated, or whether indoor temperatures were maintained at design levels. In fact, many of the homes in the dataset had shared walls and were located in mild Pacific Northwest climates, both of which naturally suppress peak heating demand.

Crucially, the Manual J models in the study were based on assumed infiltration rates—because most homes lacked blower door tests. This alone introduces a major source of uncertainty, as infiltration is one of the most variable and impactful load drivers. Without a blower door test, any load model—whether empirical or theoretical—is skating on incomplete data.

Sidebar: This is one of the strongest arguments for making blower door testing a standard part of HVAC design. Whether you're running a Manual J or trying to validate real-world usage with runtime data, measured infiltration brings clarity to what would otherwise remain educated guesswork. It also positions HVAC contractors as performance consultants, not just installers.

While Thermentor's method offers a potentially more precise look at system performance during a given season, it does not address unoccupied zones, latent load, intermittent heating schedules, or outlier weather years. Just because a system performed adequately last winter doesn't mean it's ready for this one. And that's where Manual J shines: it's predictive, not reactive.. But as the report admits, most of those homes had shared walls and mild climates. Manual J prepares for a polar vortex, not a pleasant February. And it prepares for peak humidity, not just dry-bulb averages.

The Middle Ground: Use Runtime to Validate, Not Replace
Here’s where things get interesting:

• Use tools like Thermentor to double-check Manual J outputs

• Adjust Manual J inputs based on blower door, IR scan, or data logging

• Flag outliers or obvious oversizing, but use Manual J to anchor sizing

In this way, Thermentor becomes an excellent QA tool—but not a replacement for first-principles design.

Conclusion: When the Design Matters, Manual J Still Wins
You wouldn’t build a house from a Zillow listing. And you shouldn’t size HVAC from a power bill.

When accuracy matters—for green building programs, code compliance, high-performance equipment, and occupant comfort—Manual J is still the gold standard. Runtime tools like Thermentor are promising, especially when data is solid and interpreted correctly, but they should support the process, not drive it.

Predict the extremes with Manual J. Serve the averages with Manual S. Trust the process.
Design for what’s possible, not just what’s probable.
Design with physics, not just past usage.
Design smart. Design tight.
Design with Manual J.

.