Using Whole-House Field Tests to Empirically Derive Moisture Buffering Model Inputs

Building energy simulations can be used to predict a building’s interior conditions, along with the energy use associated with keeping these conditions comfortable. These models simulate the loads on the building (e.g., internal gains, envelope heat transfer), determine the operation of the space conditioning equipment, and then calculate the building’s temperature and relative humidity (RH) throughout the year. The indoor temperature and RH are affected not only by the loads and the space conditioning equipment, but also by the capacitance of the building materials, which buffer changes in temperature and humidity. The thermal capacitance is typically included in these models, because it can strongly affect energy use. The moisture capacitance has a smaller effect on energy use, and the modeling of moisture capacitance is either simple (and inaccurate) or nonexistent in most building energy simulation programs. But this moisture capacitance has become increasingly important for modeling residential buildings because higher efficiency building codes have led to reduced sensible loads without a corresponding decrease in the moisture (latent) load. Researchers and builders are actively studying humidity control in homes, either through energy recovery ventilators, standalone dehumidifiers, or packaged air-conditioning systems with enhanced dehumidification. This research developed an empirical method to extract whole-house model inputs for use with a more accurate moisture capacitance model (the effective moisture penetration depth, or EMPD, model). The experimental approach was to subject the materials in the house to a square-wave RH profile, measure all the moisture transfer terms (e.g., infiltration, air conditioner condensate), and calculate the only unmeasured term: the moisture sorption into the materials. After validating the method with laboratory measurements, we performed the tests in a slab-on-grade house with concrete block walls at the Florida Solar Energy Center in Cocoa, Florida. We used a least-squares fit of an analytical solution to the measured moisture sorption curves to determine the three independent model parameters representing the moisture buffering potential of this house and its furnishings. After deriving these parameters, we measured the RH of the same house during tests with realistic latent and sensible loads, and then compared that to the RH predicted by the EMPD model using these inputs. This showed good agreement (Figure ES-1(a)), especially compared to the commonly used effective capacitance approach (Figure ES-1(b)). Even if we adjust the single parameter used in the effective capacitance model to try and match the data, we can only improve the R2 from 0.40 (for a commonly used effective capacitance of 10) to 0.52 (effective capacitance ~5). Both are considerably worse than the EMPD model (R2 = 0.81). These results show that the EMPD model, once the inputs are known, is an accurate moisture buffering model. A sensitivity analysis showed that the model is fairly insensitive to changes in the model inputs up to 20%. This experimental method can be used in houses of other constructions (e.g., wood frame), and with other levels of furnishings, to develop a more comprehensive dataset. This can provide guidance on moisture buffering model inputs for use in building simulations, such that the indoor RH can be predicted with greater accuracy. This can help answer questions about the effects of insulation levels, cooling equipment selection, and ventilation practices on the indoor RH, and help anticipate potential problems.

Data and Resources

Additional Info

Field Value
Author Paul Norton
Maintainer Jason Woods
Last Updated October 10, 2016, 17:59 (Etc/UTC)
Created September 21, 2016, 15:23 (Etc/UTC)
Construction Type Retrofit
Evaluation Level Test House
Climate Hot - Humid
City Cocoa
State FL
Zip Code 32922
Phase 2010 to 2015
Project Start Date 2301-02-01
Project Estimated End Date 2014-08-31
Completed House Count 1
Expected House Count 1
Average Estimated Pre-Retrofit Annual Source Energy (MMBtu) 1
Average Estimated Annual Source Energy (MMBtu) 1
Average Measured Pre-Retrofit Annual Source Energy (MMBtu) 1
Average Measured Annual Source Energy (MMBtu) 1
Average Conditioned Floor Area (sqft) 1536

Dataset extent

Map data © OpenStreetMap contributors
Tiles by OpenStreetMap