Thermal zone modeling is the practice of dividing a building into control zones that share similar occupancy, orientation, and internal loads so heating and cooling loads are calculated per zone rather than for the whole building at once. In ASHRAE Fundamentals Chapter 18 a zone is a space or group of spaces held to one uniform design condition; modeling these zones correctly is the first step of any credible load calculation and it directly controls equipment sizing, duct design, and annual energy use. Get the zoning wrong and every downstream number—coil capacity, fan airflow, chiller tons—drifts from reality.
Why Zoning Drives Load Accuracy
A single lumped load for an entire floor hides the peaks that actually size equipment. Two west-facing perimeter offices may each need 2.5 kW of cooling at 15:00 while the building core needs 1.8 kW of heating at the same hour; a single blended number would undersize the perimeter and oversize the core. Zones capture those simultaneous opposite loads so the system can be balanced, and good zoning also lets the design use part-load controls and avoid conditioning empty spaces at full design rate.
As a practical rule, group spaces that share four attributes: the same thermal envelope (wall, window, roof), the same internal-gain density, the same occupancy schedule, and the same design setpoint. When any one of those differs sharply, split the zone. A typical open office floor plate of 500–800 m² often becomes 3–6 zones, while a small retail unit of 120 m² is usually a single zone.
Two Accepted Calculation Methods
ASHRAE recognizes the Heat Balance Method (HBM) as the most rigorous and the Radiant Time Series Method (RTSM) as a simplified, widely used approximation derived from it. Both start from the same zone heat gains but treat radiant storage differently.
| Attribute | Heat Balance Method (HBM) | Radiant Time Series (RTSM) |
|---|---|---|
| Accuracy | Highest; full surface-by-surface balance | High for standard offices; weaker for atypical mass |
| Radiant delay | Simulated hour by hour | Modeled with fixed radiant time factors |
| Input effort | High; needs all construction layers | Medium; uses tabulated CTF values |
| Typical tool | Hourly simulation engines | Spreadsheet or load software |
| Best use | Complex mass, radiant floors, atria | Routine commercial load calcs |
For most commercial retrofits the RTSM gives results within 5–10% of HBM while cutting input time by more than half. Reserve full HBM for buildings with heavy thermal mass, radiant systems, or unusual geometries where the fixed radiant factors lose validity.
Internal Loads You Must Count
Zones fail when designers forget latent and sensible internal gains. Plan to include:
- People: 75–120 W sensible plus 40–60 W latent per person, depending on activity level.
- Lighting: 8–15 W/m² for LED offices, up to 20 W/m² for legacy fixtures.
- Plug and equipment loads: 10–25 W/m² in offices, 30–60 W/m² in labs or server-adjacent rooms.
- Infiltration: 0.1–0.5 air changes per hour for tight to leaky envelopes.
- Solar through glazing: up to 200–300 W/m² of south or west window at peak condition.
These values vary by use; the zone model forces you to assign them per zone instead of averaging across the building, which is exactly why zoning improves accuracy.
When to Split vs. Combine Zones
- Split when two spaces have a design temperature difference greater than about 3 °C or different schedules that offset peaks.
- Split when one space has high internal gain, such as a server closet or a kitchen, that would dominate a shared zone.
- Combine when spaces are open to each other, share a wall with no door, and run the same schedule.
- Combine small storage rooms with identical setpoints to reduce modeling effort without hurting accuracy.
- Split perimeter from core whenever the facade drives more than 20% of the zone load.
From Zones to Equipment Sizing
Once per-zone peak loads are known, sum them with a coincident factor because not every zone peaks at the same hour. A block of ten zones may show a block peak of 85 kW even though the sum of individual peaks is 110 kW, a 23% reduction that protects against oversized, short-cycling equipment. Convert cooling load to airflow at roughly 1 kW ≈ 200 m³/h for an 8–10 °C supply-air temperature difference, then select coils and fans.
Worked Zone Example
Consider a 600 m² floor with a west glass wall, an east glass wall, and an open core. Model it as four zones: west perimeter with solar gain up to 280 W/m², east perimeter, north perimeter with little glazing, and core. At 14:00 the west zone needs 3.1 kW cooling, east needs 0.4 kW, north needs 0.2 kW, and core needs 1.2 kW of heating from internal loads. The block coincident peak is 3.7 kW cooling, not the 4.9 kW sum, which guides a single 4 kW coil sized with margin rather than a 5 kW unit that would short-cycle.
Zoning for VAV and CAV Systems
Variable-air-volume systems need more zones than constant-volume because each VAV box serves one zone with its own setpoint, while a CAV system may group zones that share a schedule. In a VAV design, also avoid mixing a zone that drops below 18 °C in winter with one needing 24 °C, or the box will fight the central plant. Good zoning thus supports both load accuracy and stable control, typically reducing reheat and saving 5–15% of heating energy.
Role in Energy Simulation
Zone loads feed the energy model that predicts annual kWh and peak demand. A model built on three coarse zones can overstate cooling by 8–20% and miss the morning warm-up that drives the peak kW. Refining to six zones typically changes predicted annual cooling by 10–15% and shifts the peak hour, which changes chiller and battery sizing. Zoning is therefore not just a load step but a direct energy-model input.
Keruite manufactures this equipment at an 80,000 m² facility in Dezhou, China, with an in-house AMCA-210-style test laboratory; every unit is balanced to G2.5 and performance-tested before shipment.
Common Modeling Mistakes
The most frequent errors are using a single zone for mixed use, ignoring radiant time lag, and applying a 1.0 coincidence factor to every block. Each pushes equipment sizes high by 15–40%. Validate the model against at least one month of utility or metered data when available; a model within 10% of measured demand is generally trustworthy for design and for commissioning verification.
FAQ
What is the difference between a thermal zone and a space?
A space is a single room, while a thermal zone is one or more spaces controlled to the same temperature and served by one control point. Several identical offices can share one zone to reduce modeling work. The zone—not the room—is the unit used in load calculations and equipment sizing decisions.
How many zones should a typical office floor have?
A 500–800 m² open floor typically needs 3–6 zones: north perimeter, south perimeter, east and west if glazed, and a core zone. Small units under 150 m² may be one zone. Add zones for high-gain rooms such as server closets so their loads do not distort the general space.
Is the Radiant Time Series method accurate enough?
For standard offices and retail it is accurate within 5–10% of the full Heat Balance Method and is accepted by most codes and software. Use the full HBM instead for buildings with heavy thermal mass, radiant heating or cooling floors, or unusual geometry where the fixed radiant time factors become unreliable.
Why not just sum the peak of every zone together?
Zones rarely peak at the same hour, so summing individual peaks oversizes equipment by 15–40% and causes short cycling and poor humidity control. Applying a coincident factor based on the load-duration curve captures the true block peak and leads to right-sized, efficient, stable systems.