Comparative Evaluation of Thermal Performance of Low-Cost Housings using Alternative Building Materials under Tropical Humid Conditions

By Harimi Djamila
September 2006

The Author is a PhD Student at Sekolah Kejuruteraan dan Teknologi Maklumat of Universiti Malaysia Sabah in Kota Kinabalu, Malaysia. Her main research area is passive design, thermal comfort and energy efficiency. → See also:

Experiments were conducted at low-cost houses to evaluate thermal performance of different building materials and roofs system in East Malaysia. Considering that thermal comfort is one of the most important aspects for evaluating the impact of building materials on indoor temperature, comfort conditions were assessed using predictive formulae based on outdoor climatic data for the free running building. Thermal performance of each roof system was assessed and the comfort level of each house was discussed. Recommendations regarding building materials and roof design more suitable under hot humid conditions were suggested.

Keywords: Building Materials, Thermal Comfort, Attic Temperature, Roof Design

Introduction

Throughout the world, from ancient times, people have used passive techniques that have evolved through generations without active cooling (Jayasinghe et al. 1999). Early Malay houses can be described as raised on timber stilts and made of materials which were easily available from the tropical forests such as timber, bamboo, rattan and leaves (Ghafar 2003). Malay architecture has been modified to adapt with new materials for their durability such as concrete and bricks for walls, metal, concrete and clay tiles for roofing in several types, colours and styles. In fact most of imported building technologies are not thermally suitable under Malaysian climate. A random survey of terrace houses in Malaysia revealed that terrace houses are considered too hot by their owners half the day of the year (Davis et al., 2003). The problem of overheating is not restricted only on terrace houses (Abdu; Malik 1994; Tinker et al. 2003). To clarify the effect of climatic conditions in relation to building design and comfort conditions, Indoor and outdoor temperatures of three houses were monitored over a period of time to predict and to evaluate the indoor comfort temperature of low cost housing under Malaysian Climate. This study also focuses on possible strategies of improving indoor thermal comfort of low cost housing.

Description of the Monitoring Houses

The selected houses are located in east Malaysia, Kota Kinabalu city (the capital of the Sabah state of Malaysia). The houses are built with different building materials. Altogether, 3 inhabited houses were evaluated: OPS house, Idaman house and Serikayan house. The building material used are given in Table 1.

Table 1   Description of Materials Used
House Final Built Area Wall Roof Attic
OPS House 58.68 m² OPS Concrete (light weight - concrete) Metal Highly ventilated. 1/13 net free ventilating area
Idaman House 69.39 m² Bricks and UAC board for wall repartition Metal insulated with foil aluminum Sealed attic
Serikayan House 79.17 m² 2mm UAC board Metal Sealed attic

Data Monitoring

Data-loggers of the kind „HOBO thermometer” well protected from solar radiation and winds were fixed in the centre of the house, in the attic and outside the house to monitor the temperature day and night with an interval time of 1 hour. The temperatures were recorded from 9 July to 5 August 2004 in OPS, Idaman and Serikayan houses.

Predictive Method for Estimating Indoor Thermal Comfort

Humphreys proposed the following Eqn.1 for calculating the indoor comfort temperature for free running building based on outdoor monthly mean temperature (Humphreys and Nicol, 1998):

Tc = 11.9 + 0.534Tmmot

Where, Tc is predicted Indoor comfort temperature (°C), Tmmot is monthly mean outdoor temperature (°C).

An analysis of the climatic data of 35 years of Kota Kinabalu with Humphreys model shows that the monthly mean temperature was above the neutral comfort temperature but within the comfort limit which is ±2°C from the neutral temperature. The neutral comfort temperature in Kota Kinabalu varied from 26.1 to 26.9°C.

Auliciems (1984) tried to analyse Humphrey’s data by including some recent studies and excluding some of incompatible Humphreys data (Auliciems, 1984, Feriadi and Wong, 2004). The Eqn.2 for Comfort temperature suggested with active and passive climate control.

TC = 0.48 x Ti + 0.14 x Tmmot

Where, Ti is monthly mean indoor temperature (°C), Tmmot is monthly mean outdoor temperature (°C). This formula was used to assess the indoor comfort temperature.

Results and Discussion

Temperature Patterns in the OPS House

Figure 2 shows the recorded indoor, mid attic and outdoor temperatures. The recorded outdoor temperature during this period varied from 22.5 to 36.6°C, while the attic temperatures varied from 22.9 to 35.3 °C and the indoor temperatures from 26.3 to 31.1 °C.

Figure 2: Temperature Patterns in the OPS House
Figure 2

The indoor temperatures were always lower than the attic temperatures during the daytime and maintained well bellow the outdoor temperatures but the situation was reversed at night. The average maxima indoor temperature was lowered below the average maxima outdoor temperature by 3.7°C and below the average maxima attic temperature by 3.3°C. The average minima indoor temperature was above the average minima outdoor temperature by 3.4°C and above the average minima attic temperature by about 2.6°C.

The attic temperatures were lower than the outdoor temperatures during daytime from about 12a.m. till 5p.m. due to absorbed heat by gable walls. The Attic temperatures were higher than the outdoor temperatures during night time due to the daily heat storage dissipated from gable walls built with concrete hollow block, which increased attic temperatures above the outdoor temperatures. The system of ventilation is judiciously chosen for this house, which reduces the thermal heat transmitted through roofing material during daytime. From the statistics of the hourly indoor temperatures during 28 days analysed with Hymphreys comfort model using excel formula, it was observed that about 73.1% of the hourly data, the OPS house was thermally comfortable and about 63.7% with Auliciems comfort model.

Temperature Patterns in the Idaman House

Figure 3 shows indoor, mid attic and outdoor temperature recorded for 28 days, from 9 July to 5 August 2004. The outdoor temperatures recorded during this period varied from 22.5 to 37.9°C, while the attic temperatures varied from 23.2 to 44.9 °C and the indoor temperatures from 26.4 to 33.6°C.

Figure 3: Temperature Patterns in the Idaman House
Figure 3

The attic temperatures in the Idaman house as can be seen in Figure 2 were higher than the outdoor temperatures during daytime with slight variation at night, whereas, the attic temperatures in the OPS house followed closely the outdoor temperatures during daytime with slight variation at night which seems to perform thermally better than the Idaman house. The indoor temperature in the Idaman house could reach up to 33.59°C and seems to be slightly higher than the indoor temperatures expected with houses realised with bricks. Most probably, this is due to the lightweight wall partitions used in this house with negligible heat storage, which may explain the raised indoor temperature during daytime, whereas, the thermal capacity of building envelope realised with bricks has the effect of delaying the impact of external conditions on the interior of the building.

The maximum attic temperatures were always higher than the maximum outdoor and indoor temperatures, whereas the minimum indoor temperatures were higher than the minimum outdoor and attic temperatures at night and the minimum attic temperatures were higher than the minimum outdoor temperatures It is important to highlight that the outdoor temperatures recorded near the OPS house were slightly lower than that recorded closer to the Idaman house and the maximum difference in outdoor temperature between these two sites was about 1.31°C, whereas, the maximum difference in attic temperature between OPS and Idaman houses was about 9.62°C. The predicted indoor comfort temperature in the Idaman house with Humphreys and Auliciems models gave nearly the same results. The indoor comfort temperature in the Idaman house was acceptable for about 50.89% of the hourly data.

Temperature Patterns in the Serikayan House

The Serikayan house is located near the idaman house. The outdoor temperatures data were the same. The main characteristic of this house is the material used for walls with low thermal mass. The recorded temperatures are plotted in Figure 4 for 28 days from 9 July to 5 August 2004

Figure: 4: Temperature Patterns in the the Serikayan House
Figure 4

The outdoor temperatures recorded during this period varied from 22.48 to 37.88°C, while the attic temperatures varied from 21.33°C to 54.13°C and the indoor temperatures from 25.56°C to 37.44°C. From Figure 7.13, it can be seen that the attic temperatures during the night were lower than the outdoor temperatures.This is because the gable walls realised with a low thermal mass has negligible heat storage, and the metal roofing cooled down quickly acting as an effective nocturnal radiator. The indoor temperatures in Serikayan house were generally higher than the outdoor temperatures during night time and maintained bellow the attic temperatures during daytime.

The average maxima indoor temperature was below the average maxima outdoor temperature by about 0.18°C and below the average maxima attic temperature by about 13.63°C. The average minima indoor temperature was above the average minima outdoor temperature by about 2.31°C and above the average minima attic temperature by about 3.36°C. The Absorption and the release of heat in this house were rapid unlike the two above houses of the masonry and concrete type where absorption and release of heat were gradual. Consequently, the maximum difference in temperature of Serikayan attic was higher than that of the Idaman attic by about 7.32°C. The predicted indoor comfort temperatures range in the Serikayan house with Humphrey model was nearly similar to that predicted with Auliciems model.. The indoor temperatures were acceptable for about 52.67% of the hourly data predicted with Humphreys model and about 57.40% with Auliciems model. The maximum indoor temperatures in the Serikayan house were higher than in the Idaman house; on the contrary, the percentage of hourly indoor temperatures in the Serikayan house in terms of thermal comfort was considerable compared with the Adaman house. The Ventilation could well improuve the indoor thermal comfort in the Idaman house to an acceptable level, whereas it may not be a sufficient strategy in the Serikayan house.

Conclusion

This study has brought attention to practical solutions involved in the use of the building material under Malaysia climate. It appears that low-cost houses realised with concrete or brick walls for daytime activities such as kitchen, living and dinning room located on ground floor and wooden or light weight walls for night time activities mainly bedrooms located on first floor covered with metal roofing without ceiling could be more appropriate with regards to the available building material in Sabah.. On the other hand, in rural areas well surrounded by plants and sufficient number of trees or by a canopy of coconut palms high above the house, wooden house covered with ventilated metal roofing could be an acceptable solution under such microclimate.

References

***

Copyright © 2006, ECO Services International