Fundamentals of Humidity and Its Impact on Indoor Air Quality

Indoor air quality (IAQ) is a multidimensional concern that directly affects human health, comfort, and productivity. Among the many factors that influence IAQ, humidity stands out as a particularly complex variable. Humidity governs not only thermal comfort but also the survival and proliferation of biological contaminants such as mold, bacteria, and dust mites. Understanding and controlling humidity is therefore essential for any ventilation or HVAC design strategy.

Humidity is typically expressed in two ways: relative humidity (RH) and absolute humidity (AH). Relative humidity measures the amount of water vapor in the air relative to the maximum amount the air can hold at a given temperature, while absolute humidity is the mass of water vapor per unit volume of air. High RH levels—above 60–70%—can promote condensation on cool surfaces, leading to dampness and mold growth. Conversely, low RH levels—below 30–40%—can dry out mucous membranes, increase the risk of respiratory infections, and cause static electricity buildup. The American Society of Heating, Refrigerating and Air-Conditioning Engineers (ASHRAE) recommends maintaining indoor RH between 30% and 60% for optimal comfort and health.

Managing humidity is not simply a matter of adding or removing moisture. It requires careful integration of ventilation rates, air distribution patterns, temperature control, and often dehumidification or humidification equipment. This is where computational fluid dynamics (CFD) becomes an invaluable tool. CFD allows engineers to simulate the coupled transport of heat, air, and moisture within indoor spaces, providing detailed spatial and temporal information that cannot be obtained from simplified analytical models or experimental measurements alone.

The Role of Computational Fluid Dynamics in IAQ Analysis

Computational fluid dynamics uses numerical methods to solve the governing equations of fluid flow—namely, the Navier-Stokes equations for momentum, the continuity equation for mass, and the energy equation for heat transfer. For humidity modeling, an additional species transport equation is solved to track the concentration of water vapor. ANSYS Fluent is one of the most widely used CFD solvers for indoor air quality applications because it offers robust turbulence models, multiphase capabilities, and extensive customization options through user-defined functions (UDFs).

When applied to IAQ studies, CFD simulations can answer questions such as: How does the position of supply diffusers affect moisture distribution? Will condensation form on cold window surfaces? How do occupancy patterns and internal moisture sources (people, cooking, plants) influence local humidity levels? By answering these questions with high spatial resolution, engineers can design ventilation systems that not only meet ASHRAE standards but also minimize energy consumption and mitigate the risk of moisture-related damage.

Key Physical Mechanisms in Humidity Modeling

Moisture transport in indoor air involves several physical mechanisms that must be accurately represented in a CFD model. The primary mechanisms are advection (bulk air movement carrying water vapor), diffusion (molecular transport down concentration gradients), and phase change (evaporation or condensation of water). In many indoor scenarios, advection dominates, but near cold surfaces or in stagnant zones, diffusion and condensation become critical.

Species Transport for Water Vapor

In ANSYS Fluent, water vapor is treated as a passive scalar species that is transported by the airflow. The user specifies the mass fraction of water vapor at inlets, or alternatively, the relative humidity and temperature, from which Fluent calculates the mass fraction. The species transport equation includes terms for convection, diffusion, and any user-defined source terms (e.g., moisture addition from occupants). The diffusion coefficient for water vapor in air can be set as a constant or calculated using kinetic theory. For accurate results at varying temperatures, the variable composition of air must be considered, but for most IAQ studies, assuming constant properties with a diluted species approximation is sufficient.

Modeling Condensation and Evaporation

Condensation occurs when moist air comes into contact with a surface that is below the dew point temperature. In CFD, this can be modeled either by imposing a boundary condition that fixes the water vapor concentration at the saturated value at the surface temperature (thereby driving a diffusive flux of moisture to the wall) or by using more sophisticated Eulerian wall film models. ANSYS Fluent includes a built-in condensation and evaporation model that tracks film thickness and heat transfer, which is particularly useful for analyzing window condensation or cooling coil performance. For non-condensing surfaces, a zero-flux condition is appropriate.

Turbulence and Its Impact on Moisture Dispersion

Indoor airflows are almost always turbulent, even at low velocities, due to obstructions, thermal plumes, and diffuser jets. Turbulence enhances mixing and accelerates the dispersion of moisture. ANSYS Fluent offers a range of turbulence models suitable for indoor flows, from the standard k-ε model (adequate for many well-mixed spaces) to the realizable k-ε and SST k-ω models for better accuracy in regions with strong separation or buoyancy. Large Eddy Simulation (LES) can provide the highest fidelity but is computationally expensive and typically reserved for research or validation studies. The choice of turbulence model must be balanced against the available computational resources and the specific goals of the simulation.

Setting Up a Humidity Simulation in ANSYS Fluent

A well-structured simulation workflow is essential for generating meaningful results. The following steps outline a typical methodology for modeling humidity in indoor spaces using ANSYS Fluent.

1. Geometry and Mesh Generation

The first task is to create a computational geometry that accurately represents the indoor space, including walls, windows, furniture, and HVAC components. ANSYS DesignModeler or SpaceClaim can be used for this purpose. The mesh—the discretization of the geometry into cells—must be fine enough to resolve boundary layers near walls and diffusers, where gradients of velocity, temperature, and humidity are largest. Hybrid meshes combining tetrahedral and prism layers are common. A mesh independence study should be performed to ensure that the solution does not change significantly with further refinement. Typical cell counts for room-scale simulations range from 500,000 to several million, depending on complexity.

2. Setting Boundary Conditions

Boundary conditions define the airflow and moisture sources and sinks. Common types include:

  • Inlet (velocity inlet or mass flow inlet): Specify air velocity, temperature, and water vapor mass fraction (or relative humidity) at supply diffusers.
  • Outlet (pressure outlet): Apply ambient pressure; backflow conditions should be set with representative temperature and humidity to avoid convergence issues.
  • Wall: Assign thermal boundary conditions (fixed temperature or heat flux) and moisture boundary conditions (zero flux for non-condensing surfaces, or saturation condition for condensation).
  • Internal heat and moisture sources: Use volumetric source terms or user-defined functions to represent heat and moisture generation from occupants, equipment, or lighting.

3. Selecting the Physical Models

In ANSYS Fluent, enable the energy equation and the species transport model. Select the appropriate turbulence model (e.g., realizable k-ε with enhanced wall treatment). If condensation or evaporation is expected, enable the Eulerian wall film model or the built-in condensation/evaporation model under multiphase options. For buoyancy-driven flows (e.g., displacement ventilation), ensure that the body force-weighted scheme is used for pressure interpolation and that the gravitational acceleration vector is correctly defined.

4. Solver Settings and Convergence

Use the pressure-based solver (segregated algorithm) since the flows are incompressible at low Mach numbers. The SIMPLE or SIMPLEC scheme works well for steady-state simulations. For transient simulations, the PISO algorithm is recommended. Set under-relaxation factors carefully—typical values for moisture species mass fraction may need to be reduced to 0.7–0.9 to avoid oscillations. Convergence criteria should be based on scaled residuals (1e-4 for continuity and momentum, 1e-6 for energy and species) and on monitoring integrated quantities such as average room humidity or condensation rate.

5. Post-Processing and Analysis

Once the solution converges, ANSYS Fluent’s post-processing tools (or CFD-Post) can generate contour plots of relative humidity, temperature, and air velocity on planes of interest. Streamlines and pathlines help visualize the airflow pattern and moisture transport. Quantities such as the average RH in the occupied zone, the percentage of area above 70% RH, and the condensation mass flux on cold surfaces can be extracted for quantitative comparison. Parametric studies—varying diffuser location, supply temperature, or airflow rate—can be automated using Fluent’s journaling or Workbench parameter management.

Example Application: Optimizing Ventilation in an Office Conference Room

Consider a typical conference room measuring 8 m × 6 m × 3 m with a ceiling-mounted mixing ventilation system. The room is occupied by 12 people, each generating approximately 75 W of sensible heat and 55 g/h of moisture (mostly through respiration and perspiration). An external wall with a large window experiences outdoor temperatures and can reach surface temperatures as low as 14°C in winter, well below the indoor dew point. Without proper ventilation design, condensation forms on the window, leading to water damage and mold growth.

A CFD simulation is set up using the methodology described above. The supply diffusers deliver 150 L/s of air at 18°C and 50% RH. The exhaust is located near the ceiling on the opposite wall. The turbulence model is realizable k-ε, and the species mass fraction of water vapor is solved. Figure 1 shows the predicted relative humidity distribution at a height of 1.2 m (occupant breathing zone). The results indicate that the area near the window has RH values exceeding 75%, and the condensation rate on the window surface is predicted to be 0.2 g/m²·h—sufficient to sustain mold growth over time.

Using the simulation insights, the engineer modifies the design: a heated glazing unit is specified to keep the window surface above 16°C, and a small supplementary diffuser is added to direct drier supply air along the window. The revised simulation shows that the condensation rate drops to negligible levels and the humidity distribution becomes more uniform, with all occupied zone locations staying below 60% RH. The energy penalty for heating the glazing is offset by a slight reduction in total ventilation airflow, as the improved distribution allows the same IAQ performance with less outdoor air. This type of iterative optimization would be prohibitively expensive and time-consuming with physical prototyping alone.

Advanced Considerations and Modeling Techniques

For more complex or high-stakes applications, additional modeling capabilities can be employed.

Multiphase Modeling for Mist or Fog

When the air becomes supersaturated (e.g., in showers, swimming pools, or industrial humidification processes), water vapor may condense into a fog of fine droplets. The Eulerian wall film model does not account for airborne droplets. In such cases, a full multiphase model such as the Eulerian-Eulerian or Lagrangian discrete phase model (DPM) is needed. The DPM tracks individual particle trajectories and can model droplet evaporation, coalescence, and wall impingement. This is computationally intensive but provides the most realistic representation of two-phase flow.

Coupling with Building Energy Simulation (BES)

CFD models typically simulate the air domain at a high spatial resolution but for short time periods. To analyze seasonal or yearly performance, CFD can be coupled with a building energy simulation tool such as EnergyPlus or TRNSYS. The BES tool provides boundary conditions (wall temperatures, ventilation schedules, internal gains) to the CFD model, which returns refined heat and mass transfer coefficients. This coupled approach is particularly effective for designing natural ventilation strategies, earth-to-air heat exchangers, and radiant cooling systems where moisture transport interacts strongly with building fabric.

Using User-Defined Functions (UDFs)

ANSYS Fluent’s UDF capability allows custom source terms, boundary conditions, or transport equations. For IAQ, a UDF might model the moisture buffering effect of porous materials (e.g., gypsum board, wood), which absorb and release moisture based on local humidity. Another UDF could implement a mold growth risk index that is calculated at each wall face and logged over time. While complex to write and validate, UDFs extend the scope of standard Fluent models to address research-specific needs.

Benefits and Limitations of CFD for Humidity Modeling

The advantages of using CFD for IAQ design have been amply demonstrated: high spatial resolution, ability to test numerous design variants, simultaneous prediction of thermal and moisture fields, and visualization of condensation risk. However, practitioners must also recognize the limitations. CFD simulations require significant computational resources—a single steady-state simulation of a moderately sized room may take hours on a high-performance workstation, and transient simulations can run for days. The accuracy of results depends heavily on the quality of the mesh, the appropriateness of turbulence and species transport models, and the fidelity of boundary conditions. Validation against experimental data or published benchmarks is strongly recommended before using CFD for certification or regulatory compliance.

Moreover, CFD is not a substitute for physical measurements where high accuracy is required, such as for commissioning HVAC systems. It is best used as a design and optimization tool, informing decisions that are then verified through spot measurements and commissioning tests. When applied with sound engineering judgment, CFD modeling of humidity is a powerful capability that directly contributes to healthier, more comfortable, and more energy-efficient indoor environments.

External Resources and Further Reading

For those interested in deepening their understanding of humidity modeling and IAQ, the following resources are recommended:

By leveraging advanced CFD tools like ANSYS Fluent, engineers can move beyond simplistic assumptions and develop robust, evidence-based designs for indoor environments that promote health, comfort, and sustainability.