Search

Hygrothermal simulation tools are commonly used to assess the moisture performance of building envelope components. Owing to the computational costs required to complete simulations over the long-term, one approach to reduce simulation time when undertaking hygrothermal design analysis, is to select representative year(s) amongst sets of long-term climate data. To properly select these moisture reference year(s), a method is required to rank the climate years in terms of their moisture severity. To this end, several methods have been proposed in the literature, amongst which is the damage function method as reported in ASHRAE Project Report RP-1325. In this method, a stepwise regression model was developed to predict the damage function, as characterized by the RHT-index (integral of (Temperature - 0) (Relative Humidity – 70%)), in an OSB layer of a wood-framed wall as a function of several average yearly climate parameters for a North facing wall. The model was calibrated using climate and simulation data for eight cities in USA and validated for four cities comprising Winnipeg, Canada. The method was found to be the most consistent and accurate amongst all ranking methods as have been evaluated. The objective of this paper was to: (1) evaluate the ASHRAE method for several Canadian locations; (2) determine whether the original model can be recalibrated in the event it is shown to be deficient; and (3) explore the potential of improving the model using other approaches such as the Partial Least Squares Regression (PLSR). The results suggest that for some Canadian locations, as were investigated in this study, the use of the original model may not be appropriate for prediction. However, the model was shown to perform better after it was recalibrated for Canadian locations but without improvement in ranking performance. Furthermore, the damage function model can be improved by using PLSR in terms of prediction and ranking., 2022 Buildings XV International Conference, December 5-8, 2022, Clearwater Beach, Florida, United States, Series: ASHRAE Conference Proceedings

Hygrothermal simulation tools are commonly used to assess the moisture performance of building envelope components. Owing to the computational costs required to complete simulations over the long-term, one approach to reduce simulation time when undertaking hygrothermal design analysis is to select representative year(s) amongst sets of long-term climate data. To properly select these moisture reference year(s), a method is required to predict moisture performance and rank the climate years in terms of their moisture severity. To this end, several methods have been proposed in the literature, amongst which is the damage function method as reported in ASHRAE Project Report RP-1325. In this method, a stepwise regression model was developed to predict the damage function, as characterized by the RHT-index (integral of (Temperature - 0) (Relative Humidity – 70%)), in an OSB layer of a wood-framed wall as a function of several average yearly climate parameters for a North facing wall. The model was calibrated using climate and simulation data for eight cities in the USA and validated for three cities in the USA and one city in Canada (Winnipeg, MB). The method was found to be the most consistent and accurate amongst all ranking methods that have been evaluated. The objective of this paper was to: (1) evaluate the ASHRAE’s method for several Canadian locations; (2) determine whether the original model can be recalibrated in the event it is shown to be deficient; and (3) explore the potential of improving the model using other approaches such as the Partial Least Squares Regression (PLSR), the Least Absolute Shrinkage and Selection Operator (LASSO) regression, and the combination of LASSO feature selection and Support Vector Regression (SVR). The results suggest that for some Canadian locations, as were investigated in this study, the use of the original model may not be appropriate for predicting the moisture performance and ranking of climate years in terms of their moisture severity. However, the model was shown to perform better after it was recalibrated for Canadian locations but without improvement in ranking performance. Furthermore, the damage function model can be improved by using either PLSR, LASSO or SVR in terms of prediction and ranking.

Hygrothermal simulations are necessary to permit analyzing moisture performance when designing wall assemblies. To limit the number of simulations, simulations are performed on a limited number of wall orientations, usually the orientation that potentially leads to the worst moisture response, which is assumed to be the one receiving the highest amount of wind-driven rain (WDR). The objective of this work was to verify this assumption considering different wall assemblies and climate conditions. Four cardinal orientations (North, East, South and West) and the wall orientation receiving the highest amount of annual WDR, hereafter called default orientation. The study was conducted for four different wood-frame wall systems that differ by their claddings assembly: brick veneer, fiberboard, stucco, and vinyl. Eleven Canadian cities belonging to different climate zones were considered: Calgary (AB), Charlottetown (PE), Halifax (NS), Moncton (NB), Montreal (QC), Ottawa (ON), Saskatoon (SK), St. John’s (NL), Toronto (ON), Vancouver (BC) and Winnipeg (MB). Three scenarios were investigated. In the first case, it was assumed that there is no rain deposition on the exterior surface of the cladding (no WDR) and as consequence no rain infiltration (no moisture source). The second scenario assumed rain deposition on the exterior surface of the cladding, but no rain infiltration. The final scenario assumes rain deposition on the exterior surface of the cladding and rain infiltration through the deficiencies in the cladding. For this last scenario, the amount of rain that penetrates through openings and reaches the sheathing membrane was assumed to be 1% of the WDR as suggested by the ASHRAE Standard 160. The analyses were performed using the wettest year in each city selected among the 31-year historical climate data based on Moisture Index (MI) ranking. For the performance analysis, four different performance indicators were evaluated: Maximum Moisture Content (MC) value in the OSB, Average MC in the OSB, Maximum Mould Index (MoI) value in the outer layer of OSB and Average MoI value in the outer layer of OSB. The mould index values were negligible for the cases where there was no WDR and only WDR. Therefore, the mould index analysis was limited only for the third scenarios wherein the water infiltration was assumed. The results obtained with the four performance indicators were similar. Therefore, only the results obtained for the maximum moisture accumulated in the OSB were analyzed. For the scenario assuming no WDR and no moisture source, it was observed that, irrespective of the orientation, cladding and city, it is the north direction that leads to the worst moisture performance. For the second scenario assuming WDR but no water infiltration and for that assuming WDR and moisture source, it was found that the default orientation leads to highest moisture accumulation in most the cities while in the remaining cities, the maximum moisture accumulation occurred in other orientations than the default one. The critical case was Montreal where the moisture accumulated in OSB using default orientation (North) was significantly different than the maximum values obtained in the East direction. Overall, the results showed that, for the wall assemblies and cities considered in this study, the North orientation is the one that leads to the worst moisture performance when the wall is well protected against rain deposition. In the scenarios where there is rain deposition on the wall or water infiltration, the orientation receiving the highest amount of annual WDR generally leads to the worst moisture response. However, using only the total amount of WDR may not be sufficient for all the cities and an analysis of hourly distribution of WDR will be required before selecting the orientation for undertaking hygrothermal simulations

The indoor environmental conditions (temperature and relative humidity) are critical when estimating the hygrothermal performance of building envelopes. Measured data is not always available and as such they are estimated using any of the various model available in the literature. The objective of this work was to compare the influence of indoor conditions calculation methods on the hygrothermal responses and the moisture performance of wood-frame wall assemblies in different Canadian cities (Ottawa, Vancouver and Calgary) under historical climate loads. Various combinations of controlled and uncontrolled indoor temperature and relative humidity were calculated using approaches proposed in the ASHRAE Standard 160. These indoor conditions were implemented as indoor boundary conditions in the simulation of heat and moisture transfer for four different types of wood-frame wall assemblies, assuming no leakage in the vapour barrier. These wall assemblies differ by their cladding types: fiberboard, vinyl, stucco and brick. In each city, simulations were run for two years as selected from a historical climate data set based on the moisture index. The wall orientation receiving the most wind-driven rain for the second year was selected for simulations in each city. A 2-storey residential building having 7 m height above grade, located in suburban area, was considered. The amount of wind-driven rain impinging on the surface of the walls was calculated using the model developed by Straube and Burnett (2000). Material properties were taken from the NRC material property database. Water infiltration through the assembly was assumed to be 1% of the wind-driven rain as suggested by the ASHRAE Standard 160. Temperature and relative humidity of the outer and inner surfaces of OSB sheathing and gypsum board were compared amongst the indoor conditions scenarios. The mould growth risk on the same surfaces was used to compare the influence of different indoor conditions scenarios on the moisture performance of the walls. The indoor temperature profiles calculated for the case where only heating is present fluctuate between 21°C and up to 30°C, depending on the outdoor temperature. With air conditioning, the indoor temperature varies between 21 and 24°C. For the indoor relative humidity profiles, when uncontrolled (no dehumidifier), it can fluctuate between 30 and 70%. With dehumidifier, the indoor relative humidity is either constant or varies within a limited range (45 to 56% in Ottawa, 44 to 52% in Vancouver, and 35 to 41% in Calgary). During the winter and summer periods, the difference between the controlled and uncontrolled relative humidity can be more than 25%. Temperature and relative humidity on the surfaces of the gypsum board reflected that of the indoor conditions. For all the cases analyzed, the indoor conditions did not have any significant impact on the temperature and relative humidity profiles of the OSB, and consequently the mould growth risk did not differ among the different indoor conditions for all cladding analyzed in all cities. When there is no leakage in the vapour barrier, the difference in indoor conditions is reflected mainly on the gypsum panel and hardly reach the OSB panel. Future studies should consider the air leakage as the exfiltration of warm and humid indoor air during the heating season may lead to the condensation in the structure that modifies the hygrothermal response of wall components.