Research Accomplishment Reports 2007

Demand-Controlled Ventilation (DCV) for Residential Indoor Air Quality Control

D.G. Colliver
Department of Biosystems and Agricultural Engineering

 

Project Description

We wished to determine the optimal amount of air to bring into a house to maintain adequate indoor air quality while minimizing energy used for ventilation, then develop and test a prototype fan control system to adjust house ventiliation.

The net amount of outside air introduced into a house from combined sources of infiltration and mechanical ventilation is a complex relationship influenced by the type of mechanical ventilation system used. By monitoring the air exchange rates in test units with and without mechanical ventilation, we developed a novel way to estimate total air infiltration into buildings.

Previous work related flow to wind speed directly; our approach applies fundamental pressure relationships from which flow rates can be derived. We found the total air flow in/out of a test unit can be estimated from three parameters: wind speed, wind direction and fan flow. A convergence operation, added to the model to achieve flow continuity, greatly improved the model's ability to predict air infiltration from combined natural and mechanical sources (UK - Continuity model slope = 0.98 vs. conventional LBL model slope = 0.28). We expanded on Yuill's work, conducting a formal parametric analysis of the weather factor over a range of climates, leakage areas and building types. The weather factor is used to calculate the annual climate-specific index based on air exchange induced by natural forces for a house with known leakage. For controlling long-term emissions, average pollutant concentration over time is of interest. For variable ventilation, as exhibited by natural forces, the harmonic average of prior ventilation rates provides a superior assessment of indoor air quality.

Thirty-year weather data from eight cities, one from each DOE climate zone, were used to calculate hourly natural air exchange rates and associated pollutant concentration for various building types. Pollutant concentration values were translated into weather factors for time periods of interest. At a macro level, the results of the UK study closely resemble ASHRAE Standard 136. Standard 136 values are typically within 10%, and with the exception of Zone 8, always higher. Hence, Standard 136 overestimates air change rate of a generic building. The weather factor is relatively constant over a range of normalized leakage areas, especially for values of An > 0.5. As the envelope tightens, the weather factor tends to increase slightly. The weather factor can vary up to 25% by changing a single wind shelter class. Similar behavior occurred across all cities and weather years.

Year-to-year variation in the weather factor could be significant-- 20 to 30% within a given wind shelter class. Shorter time periods exhibit even greater variability. Thus, it is more appropriate to analyze data from a percentile approach to provide some level of confidence to a particular weather factor. From a design standpoint, to insure a reliable minimum level of air exchange, Lexington's annual weather factor would need to be reduced by approximately 40%.

Impact

Between one-third and one half of the cost of heating and cooling a well-insulated house can be attributed to air leaks. Indoor air quality concerns become important when buildings are built tighter to reduce these leaks in order to reduce the heating and cooling bills. This project seeks to determine the optimal amount of air to bring into the house in order to maintain adequate indoor air quality while minimizing the energy used for ventilation. The objective was to develop and test a prototype fan control system which will adjust the amount of ventilation in the house. The control is to be based upon outdoor temperatures and wind velocity.

Early in the research, the investigators determined that existing models for estimating natural air exchange or infiltration, i.e. air movement into a building induced by pressures from wind and temperature, did not adequately model the combination and interaction of infiltration and forced ventilation in residential structures. As a result, the investigators had to focus the research toward improved modeling of infiltration with the addition of mechanical ventilation.

Of the original four research areas (simulations, system calibration, test cell experiments, and full scale field application ), the project was able to largely complete the first three (simulations, calibration and experiments), and to demonstrate the concept on test cells however, additional analysis of the data is necessary before full scale house testing can be initiated.