My home is a classic 1960s split-level — the kind of layout HVAC engineers quietly curse. No doors between levels, a cold lower floor with tile, and a full masonry fireplace that sits at 50°F all winter. It’s a perfect storm of thermal mass and uncontrolled convection.

The furnace is a standard 80% PCMU-LD12N095B, but with one major upgrade: a custom gas-flow regulator that lets me modulate input between 27.7 and 67.3 cu ft/hr programmatically.

To understand what the house is actually doing, I use four temperature sensors:

• Upstairs room sensor (primary daytime zone)
• Downstairs room sensor (primary evening zone)
• Air intake sensor (return plenum)
• Air output sensor (supply plenum)

The intake/output pair gives me a real-time measurement of the heat-exchanger ΔT, which becomes the control variable for gas flow. The upstairs/downstairs sensors show how the house distributes that heat.

We live upstairs from 00:00–15:00, then downstairs from 14:00–00:00, so the heating strategy must adapt to two completely different thermal environments every day.

---

1. Higher efficiencies occur when outside temperatures are mild

When the outside temperature is mild:

• The house loses heat slowly
• The furnace can run at lower firing rates
• The ΔT across the heat exchanger naturally decreases
• The controller reduces gas flow to maintain the ΔT target
• ON cycles become long and stable

Mild days consistently show:

• Lower gas flow for the same ΔT
• Longer ON cycles
• Higher calculated furnace efficiency

The house simply isn’t fighting a steep temperature gradient.

---

2. Higher efficiency is also achieved at lower firing rates — but only when the house can afford it

As the outside temperature drops, the house loses heat faster. If the setpoint is above the natural equilibrium temperature, the furnace must run hotter to overcome that loss and still climb toward the target.

But once the target temperature is reached, the physics flip:

• The thermal mass is charged
• The envelope loss rate is predictable
• The furnace no longer needs to accelerate the temperature

This is where this system shines.

Once the target temperature is reached, the controller drops the furnace to 27.7 cu ft/hr, the lowest stable firing rate. At that point, the system enters a high-efficiency regime:

• Long ON cycles
• Minimal short-cycling
• Stable ΔT across the heat exchanger
• Predictable envelope losses

---

3. Intake and output sensors are the key to stable, efficient control

A thermostat only knows room temperature. This system knows how much heat the furnace is actually delivering.

ΔT = T_output − T_intake

This value becomes the real-time indicator of:

• Airflow
• Heat-exchanger loading
• Gas-input effectiveness
• System restrictions
• Envelope heat loss

The controller dynamically adjusts gas flow to maintain a ΔT target that depends on outside temperature:

• 10°C when outside temperatures are mild
• Up to 20°C when it’s very cold

This adaptive ΔT is what allows the system to run efficiently across wildly different conditions.

---

4. After reaching the target room temperature, the furnace can run at minimum gas flow

Once the home reaches the target temperature:

• The thermal mass is warm
• The envelope loss rate is known
• The furnace only needs to maintain

So the controller drops to 27.7 cu ft/hr, and the system simply maintains.

The intake/output sensors confirm that ΔT stays within the target band, and the controller trims gas flow as needed.

This is where efficiency peaks — especially on mild days.

---

5. The controversial one: closing vents can improve efficiency — but only with dynamic ΔT-based control

Most HVAC articles warn against closing vents. They’re not wrong — for fixed-speed, fixed-input systems.

This system is neither.

Because it measures real-time ΔT and dynamically adjusts firing rate, closing vents behaves very differently.

Here’s what actually happens:

1. Closing vents increases the temperature at the heat exchanger
2. The output sensor sees a higher T_output
3. ΔT increases
4. The controller automatically reduces gas flow to bring ΔT back into the target band
5. Lower gas flow = higher efficiency

Daily vent strategy

00:00–15:00 - All downstairs vents closed - Downstairs return closed - All upstairs vents open - Result: upstairs stays warm with minimal gas input

After 15:00 - All downstairs vents open - ~95% of upstairs vents closed - Result: downstairs warms efficiently, and upstairs stays at 69–70°F simply because heat rises

This strategy would not work on a traditional furnace. It works here because the controller continuously adapts gas flow to match airflow and thermal load.

---

Final Thoughts

The biggest surprise in this entire project is how much efficiency was hiding in the interaction between:

• The building envelope
• The furnace
• The airflow pattern
• The daily occupancy pattern
• The intake/output ΔT
• The dynamic gas-flow controller

A 60-year-old split-level with a cold masonry wall shouldn’t be outperforming “efficient homes,” but with the right data and a controller that respects physics, it can.

Efficiency isn’t a furnace spec — it’s a relationship between the building, the equipment, and the control strategy.