Available online at www.sciencedirect.com ScienceDirect

Energy Procedia 78 ( 2015 ) 2506 – 2511

6th International Building Physics Conference, IBPC 2015

Overhang effect on reducing wind-driven rain for a mid-rise building

Vincent Chiu*, Hua Ge, Ted Stathopoulos

Department of Building, Civil, and Environmental Engineering, Concordia University, 1455 De Maisonneuve Boulevard West, Montréal, Québec, Canada, H3G 1M8

Abstract

Roof overhangs are a common feature that can be used to protect the building façade from wind-driven rain (WDR) as field measurements and CFD modelling have shown. However, the quantitative evaluation of the effectiveness of overhangs in protecting the building façade requires further investigation. A six-storey mid-rise building with a flat roof located in Vancouver, British Columbia, Canada has been selected for the experimental study. The study involves the analysis of the spatial distribution of WDR on the building façade before and after the installation of a temporary retractable roof overhang. This paper presents the experimental setup, preliminary result analysis including the spatial variation of catch ratios on the building façade, and the performance assessment of a 1.2 m overhang under two rain events.

© 20152015 The The Authors. Authors. Published Published by byElsevier Elsevier Ltd. Ltd. This is an open access article under the CC BY-NC-ND license (Peer-reviewhttp://creativecommons.org/licenses/by-nc-nd/4.0/ under responsibility of the CENTRO). CONGRESSI INTERNAZIONALE SRL. Peer-review under responsibility of the CENTRO CONGRESSI INTERNAZIONALE SRL

Keywords: Wind-driven rain; Driving rain; Building; Façade; Overhang; Catch ratio; Quantification

1. Introduction

Wind-driven rain (WDR) is rain that is given a horizontal velocity component by the wind and falls obliquely. When considering the interaction between the rain and the vertical building façade, WDR can be defined as the component of the rain intensity causing rain flux through a vertical plane. WDR is an important research topic in building science because it is one of the most significant sources of moisture affecting the hygrothermal performance

* Corresponding author: Tel.: +1-514-531-3890 E-mail address: [email protected]

1876-6102 © 2015 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/). Peer-review under responsibility of the CENTRO CONGRESSI INTERNAZIONALE SRL doi: 10.1016/j.egypro.2015.11.252 Vincent Chiu et al. / Energy Procedia 78 ( 2015 ) 2506 – 2511 2507 and the durability of building façades. Excessive moisture accumulation on porous materials may lead to water penetration, freeze-thaw damage, efflorescence, cracking and façade soiling. Furthermore, water penetration may lead to the chemical breakdown of organic materials (such as wood), mould growth, and reduce the effectiveness of insulating materials, etc.[1]. Hygrothermal and durability analysis of façades requires the quantification of WDR loads. However, the WDR load is difficult to quantify due to the complex interactions between the wind and rain and the buildings which they affect. The quantity of WDR impinging on building façades is governed by several wind and rain characteristics, such as: rainfall intensity, duration and frequency of the rain event, wind speed, and wind direction. In addition, building characteristics such as: environment topology, building geometry, sheltering by surroundings, façade orientation, and location on the façade further affect the WDR load [1,2]. To date, there is limited quantitative data regarding the effectiveness of roof overhangs, however, studies show that the shape of the roof and overhang have a significant impact on the amount of WDR deposited on the building façade. Pitched roofs and overhangs protect the wall below it by shadowing and redirecting airflow. In 1995, Inculet and Surry [3] studied the influence of building geometry and architectural details such as balconies, cornices, pitched roofs, and inset corners on the wetting pattern of scaled down building models placed in a boundary layer wind tunnel. They found that cornices may be successful in protecting the top of the building façade just below the cornice. In 1996, CMHC [4] published a survey of building envelope failures in the coastal climate of British Columbia due to the large number envelope performance problems related to water penetration. The survey found that the lack of overhangs was a contributing factor to moisture damage in wood frame walls; as the width of the overhang decreased, the percentage of walls which had encountered problems increased. Hence, the size of a roof overhang correlated directly with the probability of rain related damage. In 2005, Blocken and Carmeliet [5] performed WDR measurements on a low-rise building with a combination of a flat-roof and a sloped-roof with varying overhang widths. The study found that the flat roof with a smaller overhang width caught significantly more rain than the sloped roof with a slightly larger overhang. Also, it was noticed that a mere 2 cm increase in overhang width significantly decreases the amount of WDR below it. Ge and Krpan [6] performed a large scale experimental study on various buildings in the coastal climate of British Columbia. The study revealed that low-rise and high-rise buildings with roof overhangs reduced the WDR exposure by about 4 and 1.5 times, respectively. Despite the amount of work that has already been put into WDR research, the quantitative effect of overhangs must be further supported by field measurements, CFD modelling, and wind-tunnel testing.

2. Experimental setup

2.1. Location and description of the study building

The study building is a six-storey rectangular residential building with a flat roof and a short parapet located near the Burnaby/Vancouver border in British Columbia (Figure 1a). The building sits atop an escarpment with the east façade facing the direction of the escarpment and is surrounded by 3-storey residential buildings to its north and west and a highway to its east and south (Figure 1b). The building is 39.2 m long, 15.2 m wide, and 19.8 m high. The building façades face the cardinal directions with one of the long faces facing the east, the prevailing wind direction during rain hours. It is a fairly open site within a suburban setting making it an ideal building for wind-driven rain studies.

2.2. Instrumentation and sensor locations

The equipment installed consists of a wind anemometer, a temperature and relative humidity probe, a horizontal rain gauge, and 31 customized driving rain gauges. The wind monitor measures the wind speed and the wind direction and is mounted on top of a tripod cross-arm that is 4.6 m above the mechanical room that is located on top of the main roof. The temperature and relative humidity probe measures the ambient air temperature and the corresponding relative humidity. It is mounted on the tripod’s mast and is shielded from the sun and wind by using a radiation shield. The horizontal rain gauge has a conical collection area (24.5 cm diameter) constructed of gold anodized spun aluminum. The resolution of the tipping bucket is 0.1 mm/tip and is placed on the center of the roof. The driving rain gauges are an aluminum plate-type gauge consisting of a square collection area, 30.5 cm by 30.5 cm, for a collection area of 2508 Vincent Chiu et al. / Energy Procedia 78 ( 2015 ) 2506 – 2511

930.3 cm2. The resolution of the tipping bucket is 0.06 mm/tip. There are a total of 31 driving rain gauges; 18 on the east façade, 11 on the north façade, 1 on the west façade, and 1 on the south façade. Initially, the building had no overhang but was later equipped with a temporary retractable overhang (0 to 1.2 meters) partially covering both north and east façades (Figure 1a).

(a) (b)

Figure 1: (a) Sketch of the test building showing the east (long face) and north façade with rain gauge locations and retractable overhang; (b) Satellite image of the study building and its surrounding environment.

3. Result analysis and discussion

The experiment has been divided into two phases. Phase I involves studying the spatial distribution of WDR on the building façade as it is without any overhang. Phase II involves installing a temporary retractable roof overhang on the North and East façades and studying the WDR pattern on the building façade at various overhang widths. The collection of data for Phase I (the building with no overhang), began August 16th, 2013 and ended December 2nd, 2014. The collection of data for Phase II (the building with a 1.2 m overhang), began December 2nd, 2014 and is ongoing. For Phase I, the rain events for the entire monitoring period with no overhang are compiled and the spatial distribution of WDR are analyzed. For Phase II, two rain events have been recorded so far with a 1.2 m overhang and are analyzed separately. The two rain events are from December 4th, 2014 to December 11th, 2014 and from December 16th, 2014 to December 27th, 2014.

3.1. Phase I: Spatial distribution of wind-driven rain with no overhang

As shown in Figure 2a, the wind direction during rain is predominantly from the east (90°, perpendicular to the east façade). The wind speed during rain fluctuates mostly between 0 and 4 m/s (89%) as seen in Figure 2b. The rainfall intensity is mostly light to moderate with less than 2 mm/hr the majority of the time (84%) as seen in Figure 2c. The total horizontal rainfall amount for the entire period is 2024.4 mm. The spatial distribution of WDR for Phase I is shown as catch ratios for the entire monitoring period with no overhang in Figure 3. The catch ratio is the amount of WDR collected on a wall surface divided by the total amount of horizontal rainfall over the same time period. The wind data (wind speed and wind direction), temperature, and relative humidity are gathered at 1 Hz sampling frequencies and averaged every 5 minutes. The sum of tips is registered for the horizontal rain gauge and the WDR gauges every 5 minutes. The 5 minute data is arithmetically averaged to hourly data, which is what is used to calculate the catch ratios. The results show that the amount of rain deposited on the building façade varies with location, however, a symmetrical distribution of WDR across the east façade can be seen due to the predominant wind direction blowing normal to the east wall. The classic wetting pattern can be seen from Figure 3: (1) the top corners are the most wetted Vincent Chiu et al. / Energy Procedia 78 ( 2015 ) 2506 – 2511 2509 followed by the top and side edges and (2) wetting increases from the bottom of the façade to the top and from the middle of the façade to the sides.

Frequency (%) 100 100 0°

337.5° 40 22.5°

315° 30 45° 75 95 20 292.5° 67.5°

10 50 90 270° 0 90° All Hours Cummulative Frequency (%) Frequency Cummulative 247.5° 112.5° 25 Rain Hours (%) Frequency Cummulative 85

225° 135° 202.5° 157.5° 0 80 180° >0-2 >2-4 >4-6 >6-8 >8 >0-2 >2-4 >4-6 >6-8 >8

All Hours Rain Hours Wind Speed (m/s) Rainfall Intensity (mm/hr)

(a) (b) (c)

Figure 2: (a) Prevailing wind direction; (b) Wind speed frequency; (c) Rainfall intensity frequency at the test building for the entire monitoring period with no overhang (Data from August 16, 2013 to December 27, 2014)

m

ES1 ES5 EC1 EN8 EN5 EN1 0.18 0.13 0.15 0.13 0.14 0.17 ES2 ES6 EN6 EN2 m 0.10 0.09 0.07 0.12 4.9 m

9.1 ES3 ES7 EC2 EN9 EN7 EN3 2.4 m 0.08 0.06 0.03 0.05 0.04 0.07 0.6

ES4 EN4 0.04 0.04

th Figure 3: Spatial distribution of WDR shown as catch ratio (ͷ= Rwdr/Rh) for the entire monitoring period with no overhang (August 16 , 2013 to December 2nd, 2014)

3.2. Phase II: Spatial distribution of wind-driven rain with 1.2 m overhang

The first rain event recorded during Phase II started December 4th, 2014 and ended December 11th, 2014. The meteorological data record of wind speed (U), wind direction (θ), and horizontal rainfall intensity (Rh) is shown Figure 4a. The rainfall intensity is light to moderate with less than 2 mm/hr the majority of the time (71%), 2-4 mm/hr occasionally (24%), and rarely reaching 4-8 mm/hr (5%). The total horizontal rainfall amount at the end of the rain event is 132 mm. The wind speed during rain hours is mostly in the range of 0-4 m/s (75%) and occasionally above 4 m/s (25%). The wind direction during rain hours is predominantly from the east-south-east followed by the east. The second rain event recorded started December 16th, 2014 and ended December 27th, 2014. The meteorological data record of wind speed, wind direction, and horizontal rainfall intensity is shown in Figure 4b. In comparison to the first rain event, the rainfall intensity is lighter with a rainfall intensity under 2 mm/hr the majority of the time (87%) and 2-4 mm/hr sometimes (12%). The total horizontal rainfall amount at the end of the rain event is 77.7 mm. The wind speed during rain hours is also lower than in rain event 1 with a wind speed mostly in the range of 0-4 m/s (96% of the time). The wind direction during rain hours is predominantly from the east. 2510 Vincent Chiu et al. / Energy Procedia 78 ( 2015 ) 2506 – 2511

When comparing rain events 1 and 2, event 1 has significantly higher catch ratios than event 2. Event 2 has catch ratios similar to the catch ratios of the entire monitoring period without overhang. This is due to the relatively higher rainfall intensity and wind speed during event 1. The wind speed and rainfall intensity is directly correlated to the WDR intensity at any given position with wind speed having a more significant influence on the catch ratio than rainfall intensity. It is evident when comparing the effect of overhang on catch ratios shown in Figures 5 and 6 with the same building with no overhang shown in Figure 3, that a 1.2 m overhang significantly reduces the amount of WDR deposition on the façade below it. The gauges EN1, EN5, and EN8 right below the overhang are the most protected. The 1.2 m overhang eliminates the majority of the WDR deposition on gauges EN5 and EN8 and drastically reduces the WDR deposition on the corner gauge EN1.The overhang is less effective in reducing WDR deposition on gauges EN1, EN2, EN3, and EN4 near the edge of the wall when compared to the 4 other rows of gauges under the overhang. Overhang protection increases as the gauges go from the side edge of the façade towards the center and from the top of the façade to the bottom. Assuming that the WDR deposition in rain events 1 and 2 (with a 1.2 m overhang) is symmetrical as it is for the entire monitoring period with no overhang the following comparisons will be made; when comparing the catch ratio of ES1 with EN1, there is around a 90% reduction in WDR for the gauges under the 1.2 m overhang for both rain events 1 and 2. Similarly, when comparing ES5 with EN5, there is almost a 100% reduction on the WDR deposition for both rain events. For locations 2.4 m below the roof overhang, the reduction is about 65% when comparing ES2 with EN2 for the two rain events. An 85% reduction can be seen when comparing ES6 with EN6 for the two rain events. The protection area provided by the overhang extends more than 4.9 m as indicated by the 40-50% reduction of WDR when comparing ES3 to EN3. Similarly, a 65-75% reduction of WDR can be seen when comparing ES7 to EN7.

(a) 10 300 (b) 10 300 Rh (mm/hr) Rh (mm/hr) U (m/s) U (m/s) θ (° from north) θ (° from nor th)t ) 8 240 8 240

6 180 6 180

4 120 4 120 (° from north) from (° north) from (° θ θ U (m/s), (mm/hr) Rh U (m/s), Rh (mm/hr)U (m/s), Rh 2 60 2 60

0 0 0 0 0 40 80 120 160 0 40 80 120 160 200 240 number of hours number of hours

nd Figure 4: Hourly horizontal rainfall intensity (Rh), wind speed (U), and wind direction (θ). (a) Rain event 1 with a 1.2 m overhang (December 2 , 2014 to December 11th, 2014); (b) Rain event 2 with a 1.2 m overhang (December 16th, 2014 to December 27th, 2015).

4. Concluding remarks

The preliminary results show that the amount of rain deposited on the building façade varies with location. However, a symmetrical distribution of WDR across the east façade occurs when the wind is blowing normal to the wall when there is no overhang. The classic wetting pattern is observed where: (1) the top corners are the most wetted followed by the top and side edges, (2) wetting increases from the bottom of the façade to the top and from the middle of the façade to the sides, (3) the WDR is directly correlated to wind speed at any given position. A 1.2 m overhang significantly reduces the amount of WDR deposition on the façade especially the areas right below the overhang. The protection provided by the 1.2 m overhang extends over 4.9 m. Results with various overhang widths under different rain conditions will be reported later and these field WDR measurements will provide valuable datasets to validate CFD models and provide quantitative evaluation of the effectiveness of overhangs, especially for mid- and high-rise buildings. Vincent Chiu et al. / Energy Procedia 78 ( 2015 ) 2506 – 2511 2511

4' overhang

m ES1 ES5 EC1 EN8 EN5 EN1 0.26 0.18 0.20 0.002 0.004 0.02

ES2 ES6 EN6 EN2 m

0.16 0.13 0.02 0.05

9.1 ES3 ES7 EC2 EN9 EN7 EN3 0.11 0.09 0.05 0.01 0.02 0.05

ES4 EN4 0.05 n/a

th Figure 5: Spatial distribution of WDR shown as catch ratio (ͷ= Rwdr/Rh) for rain event 1 with a 1.2 m overhang (December 4 , 2014 to December 11th, 2014)

4' overhang

m

ES1 ES5 EC1 EN8 EN5 EN1 0.21 0.14 0.16 0.000 0.000 0.02

ES2 ES6 EN6 EN2 m 0.11 0.09 0.01 0.04 9.1 ES3 ES7 EC2 EN9 EN7 EN3 0.08 0.06 0.04 0.01 0.02 0.05

4.9 m

ES4 EN4 2.4 m 0.04 n/a 0.6

rd Figure 6: Spatial distribution of WDR shown as catch ratio (ͷ= Rwdr/Rh) for rain event 2 with a 1.2 m overhang (December 23 , 2014 to December 27th, 2014)

Acknowledgements

Financial supports are received from NSERC strategic research Network for Engineered Wood-based Building Systems (NEWBuildS), Homeowner Protection Office branch of BC Housing, and the Faculty of Engineering and Computer Science of Concordia University.

References

[1] Blocken B, Carmeliet J. A review of wind-driven rain research in building science. J Wind Eng Ind Aerodyn 2004;92:1079–130. [2] Ge H, Krpan R. Field Measurement of Wind-Driven Rain on a Low-rise Building in the Coastal Climate of British Columbia. 11th Can. Conf. Build. Sci. Technol., Banff: 2007. [3] Inculet D, Surry D. Simulation of Wind Driven Rain and Wetting Patterns on Buildings. CMHC 1995. [4] Morrison Hersfield Limited. Survey of Building Envelope Failures in the Coastal Climate of British Columbia. CHMC 1996. [5] Blocken B, Carmeliet J. High-resolution wind-driven rain measurements on a low-rise building—experimental data for model development and model validation. J Wind Eng Ind Aerodyn 2005;93:905–28. [6] Ge H, Krpan R. Wind-driven Rain Study in the Coastal Climate of British Columbia. Burnaby: 2009.