Scholarly article on topic 'The Effects of Courtyards on Indoor Thermal Conditions of Chinese Shophouse in Malacca'

The Effects of Courtyards on Indoor Thermal Conditions of Chinese Shophouse in Malacca Academic research paper on "Earth and related environmental sciences"

CC BY-NC-ND
0
0
Share paper
Academic journal
Procedia Engineering
Keywords
{Courtyard / "Natural ventilation" / "Thermal comfort" / Shophouse}

Abstract of research paper on Earth and related environmental sciences, author of scientific article — Mohd Azuan Zakaria, Tetsu Kubota, Doris Hooi Chyee Toe

Abstract This paper discusses the effects of courtyards on indoor thermal environment in a traditional shophouse in Malacca, Malaysia based on the results of field measurement. The results showed that the indoor air temperatures in the living hall (ground floor) and the master bedroom (first floor) were approximately 0.3-1.7°C lower than the corresponding outdoor air temperature during daytime. The structural cooling effect in these rooms was reduced due to relatively large ventilation rates caused by the two courtyards. Meanwhile, during night-time, the indoor air temperatures in the two rooms were merely 0.8-1.9°C higher than the outdoors. The results of the thermal comfort evaluation showed that the indoor operative temperatures in both rooms exceeded the 80% upper comfortable limits over 36-47% of the measurement period. Furthermore, the smoke test and the sequential photograph observation revealed that there were three air flow patterns in and around courtyards during the measurement period.

Academic research paper on topic "The Effects of Courtyards on Indoor Thermal Conditions of Chinese Shophouse in Malacca"

Procedia Engineering

www.elsevier.com/locate/procedia

9th International Symposium on Heating, Ventilation and Air Conditioning (ISHVAC) and the 3rd International Conference on Building Energy and Environment (COBEE)

The Effects of Courtyards on Indoor Thermal Conditions of Chinese

Shophouse in Malacca

Mohd Azuan Zakariaa*, Tetsu Kubotaa, Doris Hooi Chyee Toeb

aGraduate School for International Development and Cooperation, Hiroshima University, 1-5-1 Kagamiyama, Higashi-Hiroshima, 739-8529

Hiroshima, Japan

bFaculty of Built Environment, Universiti Teknologi Malaysia, 81310 UTMSkudai, Johor, Malaysia

CrossMark

Available online at www.sciencedirect.com

ScienceDirect

Procedia Engineering 121 (2015) 468 - 476

Abstract

This paper discusses the effects of courtyards on indoor thermal environment in a traditional shophouse in Malacca, Malaysia based on the results of field measurement. The results showed that the indoor air temperatures in the living hall (ground floor) and the master bedroom (first floor) were approximately 0.3-1.7°C lower than the corresponding outdoor air temperature during daytime. The structural cooling effect in these rooms was reduced due to relatively large ventilation rates caused by the two courtyards. Meanwhile, during night-time, the indoor air temperatures in the two rooms were merely 0.8-1.9°C higher than the outdoors. The results of the thermal comfort evaluation showed that the indoor operative temperatures in both rooms exceeded the 80% upper comfortable limits over 36-47% of the measurement period. Furthermore, the smoke test and the sequential photograph observation revealed that there were three air flow patterns in and around courtyards during the measurement period.

© 2015TheAuthors. Publishedby ElsevierLtd.Thisis 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 organizing committee of ISHVAC-COBEE 2015 Keywords: Courtyard; Natural ventilation; Thermal comfort; Shophouse

Introduction

The energy consumption for space cooling has been particularly rising in growing cities of Southeast Asia, where

* Corresponding author. Tel.: +81-90-3177-7227; fax: +81-82-4246956. E-mail address: azuanz10@yahoo.com

1877-7058 © 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 organizing committee of ISHVAC-COBEE 2015 doi:10.1016/j.proeng.2015.08.1094

hot-humid conditions continue throughout the year. In this region, most of the modern urban houses are constructed of relatively large thermal mass materials such as brick and concrete. The use of large thermal mass materials along with the lack of natural ventilation in these urban houses often result in hot indoor conditions particularly during nighttime [1]. This leads to the excessive use of air-conditioning during the sleep at night. Therefore, application of passive cooling design is required wherever possible to improve the indoor thermal conditions in these modern urban houses for energy-saving. Meanwhile, it is said that vernacular buildings have been subtly crafted over generations in response to experience of conditions and use [2]. Recent researchers, therefore, attempted to extract traditional passive techniques employed in and around the vernacular buildings for applying them to the modern houses [1].

In Malaysia, the traditional Chinese shophouse is one of the fine examples of vernacular building. The origin of the Chinese shophouse can be traced back to the influx of Chinese immigrants from densely populated southern coastal provinces of China in the 19th century until World War II [3]. By the early 20th century, this urban design spread to every major town in Malaysia. Generally, the traditional Chinese shophouse is a narrow, deep-plan brick building situated in rows in relatively dense urban areas. Chinese shophouses in Malaysia have traditionally been two storeys high, with the lower floor used for trading and the upper floor for residential purposes [3]. An important feature of the shophouse is having one or more internal courtyards in each building, which are said to ensure indoor thermal comfort without relying on air-conditioning [4]. This paper discusses the effects of courtyards on indoor thermal environment in a traditional shophouse in Malacca, Malaysia based on the results of field measurement.

Methods

The field measurement was conducted in a two-storey traditional Chinese shophouse from 24th September to 1st October 2014. This shophouse is located in the core heritage zone of Malacca, Malaysia (2.2°N and 102.2°E). As shown in Fig. 1a, the shophouse is situated in the middle of the row with a narrow frontage (4.1m) and a long depth of 24.4m. The façade of the shophouse is facing towards southwest with a total floor area of 170m2. This shophouse was originally constructed between 1600s and 1800s with a strong influence of Dutch architecture. The building was restored and is currently managed by the Badan Warisan Malaysia (The Heritage of Malaysia Trust).

Fig. 1. Case study Chinese shophouse. (a) Location (source: google earth, retrieved on 23rd March 2014); (b) Front exterior view.

Fig. 2a shows floor plans of the case study shophouse. As shown, the shophouse has two deep atrium-type courtyards at the middle (CY1) and the end (CY2) of the elongated building. CY1 was originally designed from the beginning of construction, but CY2 was created accidentally during the renovation due to the collapse of ceiling. The sky view factor of CY1 was measured at 6.2% whereas that for CY2 was 3.1%. The thick walls of the shophouse (150-300mm) were made of clay brick with a finishing layer of lime plaster. The traditional clay tile (terracotta) was used for roofing without ceiling and thermal insulation, while the wooden boards were adopted for ceiling on the ground floor. Wooden boards were also used for floor on the first floor whereas that on the ground floor was tile. Windows and doors were made of wooden materials without using glazing. The floor material for CY1 was brick and terracotta tile, which were permeable. A well was located in CY1, which was shared with the adjacent shophouse. The room height of the ground floor was approximately 2.7m, while that of the first floor ranged from 2.7-4.5m, depending on the roof height. A veranda with a depth of 1.7m was located in front of the building.

G round Floor

O Indoor main measurement at 1.5 m above floor level • O utdoor measurement at 1.5 m above floor level IS Vertical distribution and air speed ®Indoor measurement and air speed at 1.5 m above

floor level XAir speed measurement

Fig. 2. (a) Floor plans of case study Chinese shophouse and (b) views of courtyards.

The shophouse is currently used as exhibition facility for tourists, which is operated during 11:00-16:00 on week days. The facility was operated during the measurement, but there were few visitors and therefore the shophouse was occupied by only one staff and 2-3 researchers during the period of measurement. The ceiling fans were used only when the rooms were occupied (one stand fan was used during the daytime in the office). Air-conditioners were not installed at all. The windows were opened during the period of operation (11:00-16:00). Air temperature and relative humidity at 1.5m above floor were measured at the centers of all the rooms (T&D TR-72). Moreover, the detailed thermal parameters including globe temperature and wind speed were additionally measured in the living hall (GF) for thermal comfort evaluation (Vaisala HMP155 and Kanomax 0965-03). In addition, the vertical distribution of air temperature in the two courtyards (in every 0.5m from the floor to the top) and the surface temperatures of roofs near the courtyards were measured, respectively (4.9-5.7m above ground floor level). Wind speed sensors were installed at the center of front and rear windows on each floor and at the top of both courtyards (4.8m (CY1) and 5.3m (CY2) above floor), respectively. Wind directions were also observed (using a simple ribbon) at all the windows by taking the sequential photographs (Recolo IR7). In addition, a smoke test (PSLaser Z-400) was conducted to investigate air flow patterns in and around the courtyards during daytime and night-time of the day (24th September). Meanwhile, outdoor air temperature, relative humidity and atmospheric pressure were recorded at the veranda (see Figure 2a), whereas a weather station (Davis Vantage Pro2) was placed on a small open space located about 500m away from the measurement site (at 4m above the ground).

Results and discussion

3.1 Detailed thermal environments in the living hall (GF) and the master bedroom (FF)

Fig. 3 shows the temporal variations of the measured air temperature and humidity in the living hall (GF) and the master bedroom (FF) during the whole measurement period (8 days) with the corresponding outdoor weather conditions. The outdoor air temperature and humidity in this figure were derived from the data measured at the veranda. As shown, the outdoor air temperature ranged from 24.4-33.9°C with the average of 28.3°C while the relative humidity ranged from 52-92%. There were a few rainy hours during the measurement. Daily global horizontal solar radiation recorded about 3341-6258 W/m2 throughout the measurement period. The average wind speeds recorded at the weather station were 1.5 m/s during daytime and 0.6 m/s during night-time. The measured indoor wind speeds ranged 0.1 to 0.4 m/s throughout the day. The prevailing wind direction was from SSW to SW during daytime and from NE to ENE during night-time, although the calm condition accounted for 34% over the measurement period particularly

at night. This means the prevailing wind blew towards the façade of the shophouse almost at right angles during the daytime.

24 2 1000 ss td £ ôrei 500

_ Li yinghal l_ (GF) Master bedroom

I. illi. il. ill liill. I III.

Rain period

27/9 28/9

Date / Time

â3-0 2.5

та 2.0

«15 ? 1.0

27/9 28/9

Date / Time

WNW W WSW

Wind speed S Calm: 34%

Mean: 1.5 m/s (daytime)

0.6 m/s (night-time)

Fig. 3. Temporal variations of the measured (a) air temperature and (b) relative and absolute humidity with the corresponding outdoor weather conditions. (c) Temporal variations of wind speeds and (d) wind rose during the whole measurement period.

Outdoor

27/9 28/9

Date / Time

As indicated in Fig. 3a, indoor air temperatures in the two rooms varied almost in parallel. The indoor air temperatures are approximately 0.3-1.7°C lower than the corresponding outdoor air temperature during daytime while they are merely 0.8-1.9°C higher than the outdoors instead at night. The diurnal air temperature range is normally narrowed compared with that of outdoor air temperature due to the structural cooling effect in a building with a high thermal mass. Nonetheless, this thermal mass effect is not as large as that observed in the previous measurement conducted in different Chinese shophouses [4]. This is probably due to relatively large ventilation rates in these rooms caused by the two courtyards. Meanwhile, absolute humidity in the living hall (GF) was generally 0.1 -2.3g/kg' higher

than that in the master bedroom (FF). This resulted in the difference of up to 10% in relative humidity between the two rooms.

24/9 25/9 26/9 27/9 28/9 29/9 30/9 1/10

Date / Time

Fig. 4. Temporal variations of indoor operative temperature in the living hall (GF) and the master bedroom (FF) with the corresponding

temperature limit for thermal comfort (at 1.5m above floor).

Fig. 4 shows the results of thermal comfort evaluation in the living hall (GF) and the master bedroom (FF). This evaluation was performed based on the Adaptive Comfort Equation (ACE) which was developed by Toe and Kubota (2013) for the use for naturally ventilated buildings in hot-humid climates [5]. The 80% upper limits of operative temperature from the ACE were determined based on the measured daily mean outdoor air temperature. Indoor operative temperatures were calculated based on the equation given by ISO 7726 (2001) [6]. As a result, the calculated operative temperatures in both rooms largely exceed the upper limits over 36-47% of the measurement period (most of the afternoon periods). This indicates that even if thermal adaptation in hot-humid climate is taken into account, the indoor thermal conditions during most of the daytime hours in these rooms are not comfortable unless the effects of wind speeds are considered further.

3.2 Vertical distributions of air temperature in the courtyards

Fig. 5 illustrates the vertical distributions of hourly averaged air temperature in the two court-yards during the fair weather days. Hourly averages in all the rooms and surface temperatures of the roofs are also indicated in the figure. In the early morning of 6:00, the lowest air temperatures were recorded at the two courtyards (26.5°C and 26.6°C), which were merely 0.6-0.7°C higher than the corresponding outdoor air temperature (Fig. 5a). Therefore, there were little vertical distributions of air temperature in the two courtyards. This relatively cool condition in the two courtyards is partly due to the inflows of cooled air from the roofs. As indicated, surface temperatures of the roofs were lower by up to 1.0°C than the outdoor air temperature mainly due to the nocturnal radiant cooling effect. The air near the roof surfaces was cooled and became heavier, thus fell into the courtyards. Meanwhile, the corresponding air temperatures in the other rooms were approximately 0.2-1.0°C higher than those in the courtyards. The air temperature in the master bedroom (FF) was only 0.7-0.8°C lower than those on the ground floor.

At 12:00 noon, the indoor air temperatures increased with the increase in outdoor air temperature (Fig. 5b). The indoor air temperatures in closed spaces such as office (ground floor) maintained lower values of 28.9°C compared with the outdoors (30.9°C). Nevertheless, air temperatures in most of the other spaces were increased to almost the same level as the outdoors (30.0-30.6°C). As described before, this is probably because of relatively high ventilation rates caused by the two courtyards. Besides, relatively large gradients are seen in the vertical distributions of air temperature in the two courtyards, particularly in CY1. Air temperatures near the upper opening of CY1 reached up to 33.6°C, which was 2.7°C higher than that at the veranda. This is probably because of the following two reasons. A possible reason is the radiant heat from the warmed surrounding surfaces of the courtyards. In fact, the sky view factor of CY1 (6.2%) was larger than that of CY2 (3.1%) as previously explained, thus the air temperatures near the upper opening were 1.4-1.5°C higher in CY1 than that in CY2. Another possible reason is the effect of warmed air from the roof surfaces. At noon, surface temperatures of roofs reached 50°C or more. Air was warmed near the roofs and increased the air temperatures at the openings of courtyards.

At 18:00, indoor and outdoor air temperatures still maintained high values of 30-31°C (Fig. 5c). Nevertheless, surface temperatures of roofs dropped to 33-34°C and therefore reduced the gradients in vertical distributions of air temperature in the two courtyards. Outdoor air temperature decreased from 18:00 on and reached 27.7°C at 0:00 (Fig. 5d). Air temperatures in the two courtyards became almost the same levels as the outdoors (27.4-27.5°C). The other indoor air temperatures were 0.6-0.9°C higher than the corresponding outdoor air temperature. The surface temperatures of roofs were reduced to 26.5-27.1°C and the differences of air temperature in their vertical distributions in the courtyards became smaller (0.3-0.8°C). This condition lasted until the sunrise of about 6:00. During night-time, air temperatures in the two courtyards were always 1.5-3.0°C lower than those in the surrounding spaces. This means that the two court-yards played a role as cooling sources for the surrounding indoor spaces at night.

Weather station: 26.2 °C Outdoor wind speed: 0.87 m/s Direction: NE

Weather station: 31.3 °C Outdoor wind speed: 1.63 m/s Direction: S/SSW

Weather station: 30.7 °C

24 26 28 30 32 34 36

24 26 28 30 32 34 36 Temperature (C) CoUrtyard 1 Living Hall Courtyard 2

24 26 28 30 32 34 36 24

Temperature ( C ) Courtyard 1 Living Hall (GF)

Weather station: 27.6 °C

24 26 28 30 32 34 36 24

Temperature ( C ) Courtyard 1 Living Hall (GF)

First floor level: 2.74 m

First floo level: 0 m

First floor level: 2.74 m

First floo level: 0 m

Outdoor

Front shop

Office

Outdoor

Front shop

Office

Courtyard 2

Outdoor

Front shop

Office

Courtyard 2

Fig. 5. Vertical temperature profiles in and around the courtyards. (a) 6:00; (b) 12:00; (c) 18:00; (d) 0:00.

3.3 Airflow patterns in the case study Chinese shophouse

The smoke test was conducted on 24th September 2014 during daytime (14:30) and night-time (22:00), respectively (Fig. 6). The smoke machine, which is normally used for stage play, generates smoke by heating a special liquid with a relatively high viscosity. The warmed smoke goes up because of the heat, but the temperature of the smoke soon reaches the same level of ambient air. Therefore, it was possible to investigate the air flow patterns by observing the movement of the smoke.

High Back

High Back wall

S-W direction

Courtyard 1 Height/Width (H/W): 1.14 Sky view factor: 6.2%

Courtyard 2 Back Height to width (H/W)> 2 Sky view factor: 3%

low air flow from the dispersion and backflow of neighbor buildi

High Back wall

S-W direction

Courtyard 2 Back Height to width ratio> 2 Sky vie

Fig. 6. Illustrations of air flow patterns in the case study Chinese shophouse. (a)Pattern 1 (daytime); (b) Pattern 2 (daytime); (c) Pattern 3 (night-time); (d) Smoke test during daytime; (e) Smoke test during night-time.

During the daytime (14:30), the average outdoor wind speed was 2.2 m/s and the prevailing wind direction was from SSW to SW, which is almost perpendicular to the façade of the shophouse. Two different patterns of air flow

were observed during this period (Fig. 6ab). During most of the daytime, the prevailing winds hit the high wall of the adjacent house (9m height) and created downward winds flowing into the CY2. As the air flow went downwards, it travelled through the living hall towards CY1, thus created the upward winds in CY1. It should be noted that this circulating air flow was generated not by the air temperature differences but by the air pressure difference by the outdoor winds. The above inflows of outdoor warm air in-creased indoor air temperatures, thus reduced the structural cooling effect. This means that if CY2 is closed, which is the original status, this circulating air flow would not have occurred and therefore indoor air temperatures would have maintained much lower values. Meanwhile, when the prevailing wind speeds increased relatively rapidly (gusty winds) during the daytime, a vortex was formed in CY1. The vortex occurred intermittently between the above air flow patterns (Pattern 1). The slow downward winds were still observed in CY2 during this period. This vortex is so-called skimming flow. According to Oke (1988), the disturbed air flow tends to occur in a high aspect ratio of urban layout at a certain speed of wind, resulting in the skimming flow (the aspect ratio (H/W) of CY1 was 1.14) [7]. Most of the inflows did not enter the surrounding spaces after creating the vortex. Therefore, air exchanges between out-doors and indoors were probably limited.

During the night-time (22:00), the weak prevailing winds of about 0.4 m/s blew from rear to front of the building (Fig. 6c). As described before, surface temperatures of the roofs were approximately 1°C lower than the ambient air temperature. It was observed that the smoke rose upwards after being released at CY1 at first and reached the rooftop level. Then, the smoke was cooled by the surrounding cooled air from the roofs and fell into the courtyard towards the ground floor. As a result, slow downward air flow was observed in CY1, though upward winds were seen in CY2. This is probably because roof areas, which created the cooled air, were generally larger in CY1 than those of CY2.

4. Conclusions

The results of the field measurement showed that the indoor air temperatures in the living hall (ground floor) and the master bedroom (first floor) were approximately 0.3-1.7°C lower than the corresponding outdoor air temperature during daytime despite having a high thermal mass structure. This was probably due to relatively large ventilation rates in these rooms caused by the two courtyards. The results of the thermal comfort evaluation showed the indoor operative temperatures in both rooms exceeded the 80% upper comfortable limits over 36-47% of the measurement period (most of the afternoon periods). Meanwhile, during night-time, the indoor air temperatures in the two rooms were merely 0.8-1.9°C higher than the outdoor mainly due to the inflows of cooled air from the roofs. The results of the smoke test revealed that there were three air flow patterns in and around the courtyards. During daytime, the circulating airflow (cross-ventilation) from the rear courtyard to the front courtyard was generated by the prevailing winds. Meanwhile, when the prevailing wind speeds increased relatively rap-idly during the daytime, a vortex was formed intermittently in the front courtyard. In contrast, it was observed that the dense cooled air above the roofs fell into the courtyard as a downward wind during the night-time. The deep atrium-type courtyard functioned as a cooling source for surrounding spaces particularly at night.

Acknowledgements

The authors would like to express sincere gratitude to the G-ecbo internship program of Hiro-shima University, Badan Warisan Malaysia and Institut Sultan Iskandar of Universiti Teknologi Malaysia (UTM). This research was supported by a grant from the LIXIL JS Foundation.

References

[1] D. H. C. Toe and T. Kubota, Comparative assessment of vernacular passive cooling tech-niques for improving indoor thermal comfort of modern

terraced houses in hot-humid cli-mate of Malaysia, Sol Energy. 114 (2015) 229-258.

[2] P. Oliver, Built to Meet Needs: Cultural Issues in Vernacular Architecture, Architecture Press, Oxford, 2006.

[3] V.F. Chen, (Ed.), The Encyclopedia of Malaysia, vol. 5. Architecture, Archipelago Press, Kuala Lumpur, 1998.

[4] T. Kubota, D.H. C. Toe, D. R. Ossen, Field Investigation of Indoor Thermal Envi-ronments in Traditional Chinese Shophouses with Courtyards in Malacca Malaysia, using field measurements and focuses on the cooling effects of courtyards, Journal of Asian Archi-tecture and Building Engineering. 13 (2014) 247-254.

[5] D.H.C. Toe and T. Kubota. Development of an adaptive thermal comfort equation for naturally ventilated buildings in hot-humid climates using ASHRAE RP-884 database, Frontiers of Architectural Research. 2 (2013) 278-291.

[6] International Standard Organization, ISO 7726, Ergonomics of the thermal environment - Instruments for measuring physical quantities. International Standard Organization, 2001.

[7] T.R. Oke, Street design and urban canopy layer climate, Energ Buildings. 11 (1988) 103-113.