Passive and Low Energy Ecotechniques: Proceedings of the Third International PLEA Conference, Mexico City, Mexico, 6–11 August 1984

By Simos Yannas

Passive and Low Energy Ecotechniques (PLEA) presents the proceedings of the Third International PLEA Conference held in Mexico City, Mexico on August 6-11, 1984. The book includes papers on state-of-the-art selected topics aimed at providing a basic knowledge; country and regional or personal monographs to continue the exchange of national information which is an established feature of PLEA; and position papers for the topic seminars. The text also presents papers on vernacular shelter and settlement; case studies of new buildings and retrofits, urban and community planning and design, photovoltaic systems implementation, cooling systems, modeling and simulation, guidelines and tools for design and planning.

• Language: English
• Publisher: Elsevier Science
• Release date: Sep 3, 2013
• ISBN: 9781483150048

Book preview

1984.

Part 1

Invited Papers

Outline

Chapter 1: PASSIVE AND LOW ENERGY RESEARCH AND DEVELOPMENT: A GLOBAL VIEW

Chapter 2: PASSIVE AND LOW ENERGY DESIGN FOR THERMAL AND VISUAL COMFORT

Chapter 3: CLIMATE DERIVED SHELTER AND SETTLEMENT

Chapter 4: CONSERVATION PRACTICES AND POTENTIAL IN VARIOUS CLIMATES

Chapter 5: A DECADE OF PASSIVE COOLING: A Perspective from the U.S.A.

Chapter 6: PASSIVE AND LOW-ENERGY RESEARCH AND PRACTICES-DEHUMIDIFICATION

Chapter 7: GUIDELINES FOR ENERGY OPTIMIZATION THROUGH LANDSCAPE ARCHITECTURE IN OVERHEATED ARID REGIONS

Chapter 8: EXPERIENCES AND EXPECTATIONS OF EARTH COUPLED BUILDINGS RESEARCH

Chapter 9: ILLUMINATION OF BUILDINGS AND URBAN AREAS

Chapter 10: ENERGY-EFFICIENT ATRIUM DESIGN

Chapter 11: DESIGN GUIDELINES ON VERTICAL AIRFLOW IN BUILDINGS AND URBAN AREAS

Chapter 12: WELL TEMPERED AND ILLUMINATED INTERIORS

Chapter 13: BUILDINGS, ENERGY AND URBAN MORPHOLOGY

Chapter 14: AN OVERVIEW OF THE PRESENT STATUS AND FUTURE POTENTIAL OF COMPUTERIZED ENERGY ANALYSIS

Chapter 15: THE DESIGN OF URBAN OUTDOOR SPACE: A BIOCLIMATIC APPROACH

Chapter 16: PHOTOVOLTAICS STATE-OF-THE-ART

Chapter 17: ENERGY FROM WIND IN RURAL AND URBAN COMMUNITIES

Chapter 18: RECYCLING AND METHANE PRODUCTION FOR HUMAN USE

Chapter 19: PERSPECTIVE OF BIOCLIMATIC ARCHITECTURE IN BRAZIL

Chapter 20: COUNTRY/REGIONAL MONOGRAPH INDIA

Chapter 21: COUNTRY MONOGRAPH PAKISTAN

Chapter 22: PASSIVE AND ACTIVE BUILDINGS IN THE ARABIAN GULF AREA

Chapter 23: REGIONAL MONOGRAPH - AUSTRALIA

Chapter 24: THE UNITED KINGDOM A COUNTRY MONOGRAPH

Chapter 25: PASSIVE SOLAR AND LOW ENERGY ECOTECHNIQUES IN SWITZERLAND A REGIONAL MONOGRAPH

Chapter 26: REGIONAL MONOGRAPH FOR DENMARK ON SOLAR ARCHITECTURE

Chapter 27: ITALIAN REGIONAL MONOGRAPH

Chapter 28: PRESENT STATUS OF PASSIVE SOLAR TECHNOLOGY UTILIZATION IN KOREA A REGIONAL MONOGRAPH OF THE REPUBLIC OF KOREA

Chapter 29: 100% NATURAL THERMAL CONTROL – PLUS

Chapter 30: THE TROMBE-MICHEL SOLAR WALL AND ITS PRESENT IMPLICATIONS

Chapter 31: SOLAR ENERGY STORAGE

Chapter 32: ON THE INADEQUACY OF SOLAR DESIGN TOOLS

Chapter 33: IMPROVING URBAN ENVIRONMENTS: URBAN FORESTRY AS URBAN IMPROVEMENT

Chapter 34: The Study and Use of Natural Air Flow in Buildings

Chapter 35: PASSIVE AND HYBRID COOLING SYSTEMS - A POSITION PAPER

Chapter 36: MECHANICAL SYSTEMS WITH LOW ENERGY REQUIREMENTS

Chapter 1

PLEA State-of-the-Art

Outline

Chapter 1: PASSIVE AND LOW ENERGY RESEARCH AND DEVELOPMENT: A GLOBAL VIEW

Chapter 2: PASSIVE AND LOW ENERGY DESIGN FOR THERMAL AND VISUAL COMFORT

Chapter 3: CLIMATE DERIVED SHELTER AND SETTLEMENT

Chapter 4: CONSERVATION PRACTICES AND POTENTIAL IN VARIOUS CLIMATES

Chapter 5: A DECADE OF PASSIVE COOLING: A Perspective from the U.S.A.

Chapter 6: PASSIVE AND LOW-ENERGY RESEARCH AND PRACTICES-DEHUMIDIFICATION

Chapter 7: GUIDELINES FOR ENERGY OPTIMIZATION THROUGH LANDSCAPE ARCHITECTURE IN OVERHEATED ARID REGIONS

Chapter 8: EXPERIENCES AND EXPECTATIONS OF EARTH COUPLED BUILDINGS RESEARCH

Chapter 9: ILLUMINATION OF BUILDINGS AND URBAN AREAS

Chapter 10: ENERGY-EFFICIENT ATRIUM DESIGN

Chapter 11: DESIGN GUIDELINES ON VERTICAL AIRFLOW IN BUILDINGS AND URBAN AREAS

Chapter 12: WELL TEMPERED AND ILLUMINATED INTERIORS

Chapter 13: BUILDINGS, ENERGY AND URBAN MORPHOLOGY

Chapter 14: AN OVERVIEW OF THE PRESENT STATUS AND FUTURE POTENTIAL OF COMPUTERIZED ENERGY ANALYSIS

Chapter 15: THE DESIGN OF URBAN OUTDOOR SPACE: A BIOCLIMATIC APPROACH

PASSIVE AND LOW ENERGY RESEARCH AND DEVELOPMENT: A GLOBAL VIEW*

J. Douglas Balcomb,     Los Alamos National Laboratory, Los Alamos, New Mexico 87545

ABSTRACT

Passive and low energy applications in buildings has become a topic of worldwide interest within the last few years. It has now been demonstrated very clearly that indoor comfort can be maintained with an expenditure of only 10 to 20% of the energy often required by modern buildings. This is accomplished through a combination of conservation measures to minimize the load, passive use of solar energy for heating, natural cooling, and daylighting. Hybrid systems to assist air flow or evaporation are often effective. Accompanying these developments has been an increasing level of research activity. The major research emphasis has been on devising mathematical models to characterize heat flow within buildings, on the validation of these models by comparison with test results, and on the subsequent use of the models to investigate the influence of both design parameters and weather on system performance. The results have clearly shown that correct design is very climate dependent, and many factors must be considered to obtain the best results. To this end, design guidelines have been developed, and simplified methods of analysis have been promulgated. The initial emphasis was on estimating and maximizing energy savings. Now that this is well established, the emphasis is shifting to maximizing the comfort and convenience characteristics of the building, which, when well designed, normally exceed those of conventional buildings. Performance has been monitored in test modules, test buildings, and many residential and commercial buildings. The results both confirm good performance and establish the accuracy of model predictions. A significant change in the research picture has been seen in the last 4 years; whereas the major effort was originally in the United States, research is now being conducted in many countries throughout the world as many people have realized that passive and low energy methods are appropriate in virtually every climate and are well suited to economic, convenient, and reliable building construction and operation.

KEYWORDS

Passive systems

solar energy

monitored performance

natural cooling

simulation analysis

INTRODUCTION

Passive and low energy systems have a number of characteristics that have brought them to the attention of a world now acutely aware of energy scarcity and high energy prices. These attributes include the following:

(1) Zero or minimal energy use. Passive systems rely entirely on the natural mechanisms of conduction, convection, radiation, and evaporation. Other nonpassive or hybrid systems use a minimum of external energy to run such devices as a low-pressure fan, a solar system circulating pump, or a water pump to wet evaporative cooler pads. If external energy is used, the coefficient of performance should be at least 10; that is, 10 times as much energy should be transported as input energy required.

(2) Simple and reliable operation. Passive systems usually are built as an integral part of the structure using ordinary building elements such as bricks, concrete, and glass. As such they are well understood by the occupants and, if maintenance is ever required, no special skills are needed. Low energy systems tend to use simple mechanical devices; infrequent repairs can usually be done with locally available parts and skills. The incredible complexity and poor reliability of some of the early active solar systems caused a reaction in favor of passive approaches and led to an emphasis on keeping the system simple.

(3) Low cost and multiple use. These features are combined because low cost often derives from multiple use. For example, if a sunspace provides a valuable working or connection zone, can be used as a greenhouse, and is aesthetically attractive, then most of its cost can be allocated to these functions and the energy benefits are nearly free. Perhaps the best example is a simple window which (when properly located) provides view, daylight, ventilation, and passive solar gains at appropriate times. Low cost and simplicity are also closely related.

(4) Good performance. Monitoring has shown that these systems perform well, both in terms of energy savings and thermal comfort, provided they are well designed.

Most passive and low energy systems rely on designing the building to take advantage of the climate when it is advantageous and to protect the building from the climate when it is not. This results in the use of strategies which are highly dependent on the local climate and which require a greater sophistication on the part of the designer to be able to take advantage of energy-saving opportunities afforded by the climate. The already-difficult job of the designer becomes even more involved because a whole new set of issues and constraints must be considered.

As interest in passive and low energy systems has grown, research and development has been essential to provide both credibility and guidance to designers. The expertise of the technical and scientific community has been enlisted to evaluate and predict performance, to develop design tools, and to assist in the development of new products and methods. Initially, emphasis was placed on passive solar heating, but gradually a much broader view of building energy issues has led to a balanced consideration of conservation, winter heating, summer cooling, and daylighting. Also, an initial emphasis on single residences has broadened to include multifamily residential and both small and large commercial buildings. Whereas research and development was originally concentrated in the United States it has quickly spread to many countries around the world encompassing a wide range of climates and building traditions.

PASSIVE RESEARCH

Research on passive solar heating has been reviewed recently by Balcomb (1982) and this work will only be summarized briefly here. A schematic overview of the key elements is shown in Fig. 1. Mathematical modeling is the critical element providing a bridge between the experimental activities on the left side and the systems analysis on the right side. Thermal network or other mathematical techniques are used to characterize the heat flow and general thermal behavior.

Fig. 1 Schematic of the key elements of the research program.

Experimental Work

Test modules have proved to be especially valuable experiments. These have ranged from test boxes of about 1 m on a side to test rooms of about 13 m³ volume with 2.5 m² of glazing and to small test buildings. Test modules in the U.S., described by Moore (1982), are usually operated in actual outdoor weather conditions and provide an opportunity to obtain data, usually on a single passive solar element, under carefully observed and controlled conditions. These experiments have provided the bulk of data used for validation of the mathematical models. In most cases, relatively simple models have been adequate to predict average temperatures and back-up heat requirements. For example, combining mass elements can often be done if the only information desired is about average behavior; whereas, if detailed prediction of an individual wall temperature is needed, that wall must be considered separately in the model.

Special experiments are usually set up to obtain information about one specific energy flow mechanism. For example, similitude experiments have been performed by Weber (1978) to obtain data on natural airflow through an aperture, such as doorway between rooms. The experiments were performed in the laboratory using a one-fifth scale model filled with Freon® gas in order to achieve similitude in the important flow parameters. Heat was transfered from a hot plate at the end of one room through the aperture to a cold plate at the opposite end of the other room. Correlations were obtained for the heat exchange (in terms of a Nusselt number) as a function of the room-to-room temperature difference (in terms of a Grashof number). It was determined that the heat exchange is quite sensitive to the door height, presumably because of hot gas trapping above the door top.

Performance Monitoring

Whereas detailed results existed from only a handful of monitored buildings before 1980, there is now a huge volume of data from the monitoring of more than 100 passive solar buildings in the U.S. alone, representing a wide variety of climates and design approaches. Most of this data collection has been the result of the Class B monitoring program and, to a lesser extent, the National Solar Data Network program. Hourly data from 20 or more channels are recorded and computer analyzed to determine building energy requirements, auxiliary heat, and internal heat, usually summarized monthly. Taken as a whole, these data show very good net performance. Measured building load coefficients are usually in the range of 0.8 to 1.5 W/Cm², which is about one-half that of conventional contemporary buildings. Auxiliary heat requirements are in the range 0.25 to 0.50 W/Cm² in sunny climates and 0.4 to 0.8 W/Cm² in less favorable solar climates. Solar gains offset typically 25 to 75% of the building load. There are a wide variety of buildings in the sample including superinsulated, earth sheltered, and all the major passive system types. Internal heat varies widely between the various buildings and makes a major contribution to the heating in some cases. Although no passive type emerges as the best performer, good thermal design is seen to be essential. The energy performance of 48 of the monitored buildings is summarized in Fig. 2.

Fig. 2 Results of the monitoring of several buildings. The bars show seasonal energy (usually for 5 or 6 months) divided by the building floor area and the actual degree days for the season, calculated for a base temperature of 18.3°C (65°F). The black portion of the bar denotes purchased energy; the portion below the bar is internal energy, and the black portion above the bar is auxiliary heat. The total length of the bar is the total heat required by the building, determined using the building heat load coefficient and the measured inside/outside ΔT integral. Thus, by subtraction, the white portion of the bar is the solar energy absorbed less any vented energy. The state in which the site is located is indicated above the bar. The buildings are rank ordered according to auxiliary heat. Several buildings with low internal heat were unoccupied but were thermostatically controlled to normal levels.

A deficiency of the monitoring is that evaluation of thermal comfort has been almost overlooked in favor of energy performance measures. While most of the buildings are comfortable, there is a tendency for temperature swings to be too large in some direct gain buildings.

A separate survey by Meier (1984), which includes both passive solar and other low energy houses, finds that space heating energy has been economically reduced to about one-fifth the level required in the average existing house or about one-third the level estimated for typical new homes.

Data from monitoring are a valuable resource for many purposes. Regression analysis has been used to determine constants in simple thermal network models using the hourly monitored data. Validation of design tools, such as solar load ratio correlations, is underway. In addition to the Class B data, much more detailed data are now becoming available from the Class A monitoring program. A primary objective is detailed and comprehensive validation of computer simulation models of buildings.

Systems Analysis

Systems analysis has consisted mainly of running computer simulation models through an entire year using historical hourly weather and solar data taken from a large number of sites. The results have been used both to study the effect of various design parameters on performance and as a data base for the development of design tools.

Sensitivity to the selection of design parameters has usually been presented as a graph showing how a single parameter, such as Trombe wall thickness, affects annual performance in a particular city. The results are depicted either in terms of heating and cooling energy (as in the California Passive Handbook) or in terms of a dimensionless solar savings fraction (as in the Passive Solar Design Handbook). Since there are many parameters, the job of analyzing and presenting results becomes a large one.

A variety of passive solar design tools have emerged. Perhaps it would be more accurate to categorize these as analysis tools helpful in design. These are normally based on correlation techniques and require much less calculation than a simulation, although many are quite complex and are most suitable for computer analysis. Others are more simplified, but usually at the cost of being less comprehensive.

The solar load ratio method (Balcomb, 1982) has been widely used, in which correlations for each of 94 passive system types have been developed. Calculations are done monthly, based on long-term average temperature and solar statistics; the analysis, however, can be reduced to a much simpler annual calculation if prepared tables are available for the site of interest.

Balancing Conservation and Solar

A technique has been developed by Balcomb (1980) to determine the optimal mix between conservation and solar strategies. To obtain an answer, one needs the cost characteristics of both the passive solar system and the energy conservation features. This information will generally be in the form of the cost per R per unit area for the wall and ceiling insulation, the cost per additional glazing for windows, the cost of reducing infiltration (including the cost of adding an air-to-air heat recovery unit if needed), and the cost per unit area for the passive solar collection aperture. Given this information, the method provides simple equations that can be used to trace the economic optimal-mix line for a particular locale.

Passive Cooling

Although passive cooling (sometimes referred to as natural cooling) has received much attention since about 1978, the evolution from research results to quantified performance evaluation, design tools, and appropriate products has been much slower than for passive heating. This is partly because of the nature of the problem. For natural cooling, the building is frequently open to the atmosphere, for example, to promote natural ventilation, whereas for heating it is normally closed. Thus the problem is less tractable to simple analytical modeling because terrain, external velocity and pressure distributions, and details of building geometry become relatively much more important. A second reason is that natural cooling comprises a set of strategies which are only related in that the objective is to promote heat rejection. These are natural ventilation, radiation, earth contact, and evaporation. Nonetheless, these systems must work together. Also, it is often the case that the most important strategy is not cooling itself but the avoidance of a cooling load through strategies such as shading and light exterior colors. The need for dehumidification is often the remaining major issue when defensive strategies have been employed. Thus the problem is highly interrelated and nonlinear. The main approach has been brute force computer simulation, and it has proved to be difficult to categorize the results into a simple set of guidelines and analysis procedures.

Nonetheless, passive cooling works. Radiative cooling has been the most researched probably because it is analytically the most tractable. It works best in arid climates at night when the sky temperatures are low. Earth contact has also been well studied; although cooling can be achieved through earth contact, the amounts are small. It is most appropriate to midcontinental climates with cold winters and hot summers. The primary benefit is probably buffering the building facade from the extremes of the outside climate. An unfortunate side effect is that the opportunity for natural ventilation is reduced.

Natural ventilation, especially ventilation at night when outside air temperatures are low enough, is probably the most effective passive cooling strategy. It is also the least amenable to precise analysis and prediction. Major studies and experiments are underway to study all these effects, and more research results should be forthcoming in the next few years.

PASSIVE SOLAR RESEARCH AROUND THE WORLD

The following are a few selected examples of passive solar heating research being performed in various countries to give a flavor for the different issues being addressed and different approaches being taken.

Portugal

A demonstration and research passive solar test house has been constructed by the faculty of engineering at the University of Porto. The house, called Thermally Optimized House-Laboratory has a gross floor area of 145 m² on two levels. It is built in close accordance with contemporary construction practices and materials except that insulation is added outside the wall and roof mass to bring the overall daily loss coefficient (per unit floor area) down to 27.5 Wh/m²-C-day. The wall U-value, for example, is 0.6 W/m²-C. Passive solar collection area is 24.5 m² consisting of water wall, Trombe wall, and direct gain. These values are consistent with conservation and solar balancing guidelines developed for Porto based on the procedure referenced above.

Simulation analysis has been performed for the building, based on the thermal network model shown in Fig. 3. This indicates that good comfort conditions will be achieved and that the building will respond in a very slow and well-behaved way to normal and extreme weather transients. Total auxiliary heat is expected to be about 1400 kWh during a normal winter based on 1615 heating degree days (18°C base).

Fig. 3 Thermal network of the Porto solar house. Numbered circles show points at which temperatures are calculated. Resistor connections show heat flow paths, labeled G. The model is driven using outside temperature and solar radiation inputs.

The building is very thoroughly instrumented and a microprocessor-based system is being used for data acquisition and recording. Provision is made to transfer data to a central computer facility for subsequent analysis and model validation. The building is a very well conceived response to the social and climatic and economic needs of Portugal and the results will be quite valuable in enhancing the credibility of passive solar as well as providing valuable research results to guide evaluation of design guidelines.

Argentina

A team of architects, engineers, and physicists has been assembled at IADIZA (Argentine Institute for Arid Zone Investigations, a government-sponsored agency) in Mendoza (33 degrees S latitude). The group has been assimilating information on appropriate climatic design from around the world with the help of consultants brought in for a few weeks. A small test house incorporating Trombe walls, direct gain, and thermosiphon solar water heaters was built in 1980 and has provided valuable feedback to the group.

Argentina has a large public housing program and there are clear indications that climatic factors should be considered in the design (the houses are sometimes referred to as horintos - little ovens). The IADIZA group has investigated means of retrofitting existing houses to be better behaved thermally and has developed designs for multistory apartment buildings that will pass the demanding earthquake standards of Mendoza. The favorable weather conditions of the area indicate that very small heating and cooling loads can be achieved through good design.

A very different design challenge has been addressed by the group at the University of Salta for buildings to be located in the altiplano of north-west Argentina (22 degrees S latitude). The high altitude (3300 m) and extremely low humidity results in a very sunny but quite cold climate (3230 degree days) characterized by huge diurnal temperature swings (−7 to +15 C in midwinter) and a year-round need for heat but no need for cooling. These conditions coupled with the low latitude lead to different design strategies such as a central atrium with horizontal glazing. One building has been constructed with east- and west-facing Trombe walls in addition to normal north-facing solar collection planes.

China

Reseachers at the Gansu Natural Energy Research Institute at Lanchou in north central China are setting up a series of test buildings that include passive solar heating strategies appropriate to the cold, sunny winters of this continental climate. Favorable data have already been obtained on a modified Trombe wall house at Xinging. The particular challenge for this group is to develop approaches and materials suitable for a country where glass is not readily available in large sizes at affordable prices, and where there is no established large-scale building insulation industry. Traditional buildings are already designed in accordance with good passive solar features such as correct orientation, building shape, location of windows, window overhangs, and high thermal mass. But performance is limited by high heat losses and poor glazing performance. Results of the research will be used to guide government policy regarding materials manufacture, retrofit of existing buildings, and the design of new buildings for the 200 million people living in this climate zone.

Spain

Researchers at the Gas and Electricity, S.A., on the Mediterranean island of Mallorca have built a small test house to experiment on a hybrid heating system. Solar heat is collected in a vertical air-heating panel incorporated into the structure of the building’s south wall. The heated air is forced through ducts by a low-power fan to a central partition wall having labyrinthine channels built of ordinary brick wall tiles sandwiched between massive wall surface elements. The purpose is to develop a means of distributing heat from the south side of a building to thermal storage within north rooms, a common design concern in high-density housing. Heat distribution to the house is by passive conduction through the wall and radiation and convection to the space. Daily collection efficiencies exceeding 35% have been obtained along with very stable and comfortable room temperatures.

A related problem in high-density housing is the distribution of daylight into interior rooms located below the upper story. Architect Rafael Serra at the University of Barcelona has experimented with vertical ducts which extend from light scoops located above the roof downward into the building. The scoop and ducts are lined with mirrors to obtain high light transfer efficiency. Reasonable lighting levels can be achieved in several rooms feeding off a single duct. The aesthetic quality achieved is excellent and exciting.

European Community

Lebens (1983) has reported on a variety of design issues and concerns that have been addressed in a multinational way under the Commission of European Communities Passive Solar Program. Their activities include sponsoring two separate European Passive Solar Design Competitions, funding many component research and development projects, developing performance monitoring techniques and monitoring buildings, evaluating both simple and simulation design models, developing design guidelines, constructing test facilities, and publishing an extensive European Passive Solar Design Handbook. The overall scope of research work is huge; it is a well-coordinated effort involving nine countries having a similar set of climatic and design concerns.

United States

Research on active solar in the U.S. started gaining momentum in 1974 and in passive solar in 1978. This broad effort has involved thousands of researchers in both small and large groups located throughout the country at government laboratories, in universities, in industry and industry associations, and in private offices. Many of the results have been reported at the annual conferences and the eight passive solar conferences sponsored by the American Solar Energy Society. The proceedings of these conferences provide an access point into most of the U.S. research work.

Research on passive and low energy buildings has decreased dramatically in the U.S. since 1980 as a result of reductions in government funding, and is probably at no more than 25% of the 1980 level today. The nature of the program has shifted from an emphasis on commercialization to an emphasis on more fundamental research and development.

REFERENCES

1. Balcomb, J. D. (1980). Conservation and Solar: Working Together. Proc. 5th Passive Solar Conf., Amherst, MA. (44–48). American Solar Energy Society, Boulder.

2. Balcomb, J. D., Passive Solar Heating Research, Advances in Solar Energy; 1. American Solar Energy Society, Boulder, 1982.:265–304.

3. Balcomb, J. D., Jones, R. W. (Ed.), Kosiewicz, C. E., Lazarus, G. S., McFarland, R. D., and Wray, W. O. (1982b). Passive Solar Design Handbook, Vol. 3, U.S. Department of Energy Report DOE/CS-0127/3.

4. Lebens, R. M. (1983). The Commission of the European Communities Passive Solar Programme. Proc. 2nd Intl. PLEA Conf., 699–707. Crete, Greece.

5. Meier, A. K. (1984). Monitored Performance of New and Retrofitted Buildings. Proc. PLEA 1984 Conf., Pergamon, London.

6. Moore, E. F. and McFarland, R. D. (1982). Passive Solar Test Modules. Los Alamos National Laboratory report LA-9421-MS.

7. Weber, D. D. (1980). Similitude Modeling of Natural Convection Heat Transfer through an Aperture in Passive Solar Heated Buildings. Ph.D. Diss., Univ. of Idaho. Los Alamos National Laboratory report LA-8385-T.

---

*Work performed under the auspices of the US Department of Energy, Office of Solar Heat Technologies.

PASSIVE AND LOW ENERGY DESIGN FOR THERMAL AND VISUAL COMFORT

S.V. Szokolay,     Architectural Science Unit, University of Queensland, St. Lucia, Queensland, Australia

ABSTRACT

After outlining a passive philosophy in the introduction, the conditions of thermal comfort are examined and the variability of requirements in space and time are emphasised. The revised bioclimatic chart is introduced and the sliding scale method of analysis is suggested. For passive solar heating design the pattern-matching method is introduced and its relevance to passive cooling design is examined. The importance of psychological factors is underlined. Luminous comfort is discussed in terms of the adequacy of illuminance, suitability of vector direction and vector/scalar ratio. The difficulty of achieving luminous comfort with daylighting is examined and the PSALI method is considered. Beam sunlighting is proposed to replace PSALI.

INTRODUCTION

The given environmental conditions are often unsuitable for human existence, therefore we build shelters and install lighting, heating or cooling equipment and thereby we create a controlled environment. The task of these controls can be defined conceptually as the difference between the given environmental conditions and the required conditions.

Two sets of means are at our disposal to perform this control task:

1. passive controls: the building itself, its shape, position and fabric

2. active controls: the various installations, based on some form of energy supply.

This conference, as a whole, represents the view that the control task should be performed by passive means, as far as practicable, and active means should be employed only for any residual task:

where active controls tend towards zero.

The aims of the present paper are:

1. to attempt a definition of comfort

2. to identify the control task

3. to examine how far we can go with passive controls.

The word comfort is defined by the Oxford Dictionary as pleasure, delight, physical well-being, …the condition or quality of being comfortable, and in turn, comfortable is explained as …satisfactory, affording or fitted to give tranquil enjoyment and content. The topic I was asked to discuss is thermal and visual comfort, which narrows down and further specifies the meaning.

I particularly like the tranquil enjoyment phrase – it leads to a very practical definition in negative terms: the absence of discomfort. We are in a state of comfort, when we are unaware of the thermal and luminous conditions, when there are no disturbing thermal or luminous influences, when we can do what we want to do, without encumbrance, without effort (due to thermal or lighting conditions).

Effort is the key term. Our physiological control mechanisms can compensate for over-, or under-heated conditions, but it takes an effort, a thermal strain. We can strain our eyes at a visual task, with 1 or 2 lux illuminance, or we can squint and screw up our eyes at 100 000 lux, but it takes an effort to perform the task. Comfort exists only when no such effort is required.

THERMAL COMFORT

A fairly comprehensive review paper I wrote is soon to appear in the second volume of Advances in Solar Energy (The ASES annual publication) so I won’t repeat its contents here, only summarise the main issues.

The equation M ± Cd ± Cv ± R − E = 0

(where M = rate of metabolic heat production

Cd = conduction)

Cv = convection) heat loss or gain rate

R = radiation)

E = evaporation heat loss rate)

gives a concise expression of the body’s thermal equilibrium. Our metabolic heat production (which can vary between 70W at sleep and 700 W at very heavy work or vigorous sport) must be dissipated, thus the heat flow is always away from the body. This equilibrium is a dynamic one: the uncontrollable change of one factor can be compensated for by an inverse change of another – within limits.

The various heat flow processes have been analysed in great detail (e.g., Fanger, 1970, or McIntyre, 1980) and the study of thermophysiology became a well developed science in its own right. The physical/physiological processes however do not tell the whole story.

Following the work of Auliciems (1981) the levels of thermal perception can be summarised as in Fig. 1. This model gives a schematic representation of complex psychological processes. The thermal sensation is filtered through several levels of mental processes, before it leads to an expression of preference or judgement. The main determinants may be physical (environmental conditions, activity, clothing) but there are other influences, from the state of acclimatisation of the individual, his or her expectations and even attitudes, to behavioural adjustments. And these influences are often interconnected. To give just one example: at the ‘effective’ level a behavioural response:‘taking off the jacket’ is activated, which will have an influence at the ‘discriminatory’ level, changing the heat dissipation mechanism, thus modifying the sensation – but such a response may be prevented by socio-cultural constraints.

Fig. 1 Psychophysical model of thermoregulation showing levels of perception and response.

The feed-back loop shows short and long term responses, e.g.:

1. change clothing

2. change thermal environment: open window, switch on heater, etc.

3. slow change of acclimatisation

4. change of habits

5. long term cultural changes in living pattern, etc.

This model helps to explain the findings of Humphreys (1978) and Auliciems (1981), that ‘thermopreferendum’ varies from place to place and even from season to season, as a function of monthly mean outdoor temperature. ‘Thermal neutrality’ (Tn), i.e., the mean of preferred temperatures for a large sample, can be expressed as

Tn = 17.6 + 0.31 × To(°C)

(with probable practical limits of 18.3 and 29.5°C) where To is the monthly mean outdoor temperature.

The ‘comfort zone’, i.e., the range of acceptable temperatures can then be taken as ±2.5K about this neutrality temperature.(°C is used to denote a point on the temperature scale, whilst K = Kelvin is used to denote a length on the scale, i.e., a temperature difference or increment.)

APPLICATION

A multitude of single-figure thermal comfort indices have been produced by various research workers, but none of them is very useful in practical design work. They all tend to conceal the influence of individual environmental variables, some of which may be controllable, others not. Olgyay’s bioclimatic chart distinguishes the influence of the four main environmental variables: air temperature, humidity, radiation and air movement, whilst also indicating their interaction.

Arens (et al., 1980) proposed a revised version of this chart (Fig. 2), incorporating the results of the latest research findings. The only major argument I have with this chart concerns the wind effects at low humidities. As soon as the air temperature is higher than the skin temperature, there will be a convective heat input into the body. The evaporative cooling may continue and compensate for this, but there will be a cross-over point where the convective gain becomes greater than the evaporative heat loss, so the increased velocity will give an increased temperature sensation. At very low humidities high air velocities will tend to dry out first the mucous membranes, then the skin.(The sweating mechanism is fatiguable!)Until these issues are clarified, it seems more reasonable to retain Olgyay’s original wind-line pattern.

Fig. 2 The bioclimatic chart, as modified by Arens et al. (1980).

Fig. 3 shows the bioclimatic chart I propose for provisional acceptance. The upper and lower comfort limits are indicated at the 50% RH line by points A and B. The comfort zone is 5K wide.(Arens’ version is only 3.5 K wide, Olgyay’s original was 6.5 K wide.)Its centre is marked by point C.I propose to use a sliding temperature scale on the left, so that the neutrality temperature (calculated for the location from Auliciems’ expression) on scale ‘A’ is adjusted to the level of point C.

Fig. 3 Revised bioclimatic chart, with sliding temperature scales (based on Olgyay, 1963, and Arens et al., 1980).

For activities other than sedentary an adjustment can be made: for every 100 W increase in metabolic heat production the temperature is reduced by 2.5 K (with a maximum of 7.5 K adjustment). Four scales are shown:

A further adjustment can be made for clothing: for every change of 1 clo unit a change of 0.6 K in the opposite direction. Appendix l (after Berglund, 1980) gives a list of clo values of clothing elements and the method of finding the total clo value. This adjustment should, however, only be used in case of people wearing abnormal clothing, as the neutrality temperatures are valid NOT for a specified clothing level, but for people (fully acclimatised, at sedentary work) wearing their normal choice of clothing.

Appendix 2 shows a climate analysis of Darwin and Tennant Creek: a warm-humid and a hot-dry location in Australia, using this SLIDING SCALE METHOD.

The setting of thermal standards has obvious energy consequences for conditioned (heated or cooled) buildings. The fabric and ventilation losses (or gains) will be proportionate to the temperature differential (ΔT). For example, in the New York winter, when the To is 0°C, the following heat loss ΔT values obtain by different standards:

thus the adoption of Tn for indoor temperature would give a reduction of over 23%. Conversely, for the Las Vegas summer, when the To is 29.5°C, the heat gain AT values would be:

thus the cooling requirement would be reduced to less than half.

The use of our sliding scale method for passively controlled buildings would mean a much improved chance of success. And such success would be achieved NOT by the lowering of standards but by recognising that human beings are not machines and by matching the standards to the variable human requirements, thus increasing the probability of comfort.

PASSIVE THERMAL CONTROLS

There are a number of studies (and a wealth of anecdotal evidence) that people generally don’t like air conditioning. There appears to he a willingness to accept some degree of thermal discomfort, as long as windows can be opened and direct contact with the natural environment can be maintained. There is presumably a limit in thermal stress, when the otherwise undesirable air conditioning would be accepted.

Reaction to heating does not appear to be so adverse. One of the probable reasons for this is that heating is more transparent, obvious and comprehensible, it is well within everyone’s experience, unlike the complicated cooling machines. It is at the same time much easier to achieve heating by passive means than passive cooling.

Two environmental influences are dominant on buildings: air temperature and solar radiation. In a winter situation these two can be played against each other, as they have opposite signs. In a summer situation both have the same sign, both cause heat gain, so one has to resort to much more subtle and indirect techniques. One factor in favour of passive cooling is people’s aversion towards mechanical methods.

However, passive cooling techniques can also attract adverse reactions. We have – for example – demonstrated that even in Darwin, for at least ten months of the year, the massive building, closed off early morning after overnight ventilation, will remain significantly cooler until about sunset, than the outdoor air and very much cooler than its lightweight counterpart. Most people however, prefer to open the windows, ’to get the fresh air in’apparently even when that fresh air is much warmer.

Passive Heating

The principles of passive solar heating systems are well known, thus there is no need to repeat them here. Their performance can be characterised by two simple indicators:

i.e., the amplitude or ‘swing’ of indoor temperature changes.

. The heat gain at peak times is ‘unutilizable’. An increased thermal mass (i.e., increased heat storage capacity) may reduce these swings.

Fig. 4 Temperature profiles for three typical passive systems (after Balcomb, 1980).

The choice of system should be matched to the requirements, e.g., afternoon overheating may be acceptable in a space used at night only; or night-time drop of temperature below the comfort zone may be acceptable in a space used during the day only.

A new type of design attitude is necessary, based on the realisation that we are creating dynamic systems. Steady-state assumptions are inadequate and our mental capacity is insufficient to handle discrete quantities through their repeated variations. We must make some abstractions, comparable in simplicity to the steady-state assumptions, if we want to avoid getting bogged down in details. We must look at the pattern of variables. The design job then becomes a pattern-matching exercise. The pattern of outdoor temperature variations (pT), and the pattern of solar irradiation (pS) are given. The occupancy pattern, or use-pattern (pU) can be readily established. We have to interpose between these sets a building system, which would give the response pattern (pR) required to bridge the gap between pU and pT+pS.

As an illustration .

Fig. 5 Pattern matching (a July day in Canberra)

For a school room (8–18 h.) there is no need for phase-delay: a direct gain type system would give the required pR (1).

For a living room (16–22 h.) the choice is

;

: probably a Trombe-wall type system.

. Ideally the Ti should not be greater than 2.5 K, (i.e., 5 K from peak to peak) but we may allow it to exceed or drop below the comfort limits during non-use periods.

. These gains will cause an increase of the indoor mean temperature, consequently an outward heat flow. This outward heat flow must equal the above extra gains, from which the increase in indoor mean temperature can be found as

where q = qc + qv, the specific heat loss rate of the building:

qc = Σ(A * U) (A=area, U=transmittance of each element)

qv = 0.33 * V * N (V=volume of room, N=number of air changes per hour).

.

If the time-lag and decrement factor characteristics of each element are known, the deviation from the daily mean heat flow (Q) for any hour of the day can be readily calculated but if this calculation is to be repeated for the 24 hours, it can become lengthy and it is preferable to employ a computer. This deviation in heat gain from the mean will be either absorbed in the fabric of the building or removed by ventilation. Admittance (Y) is a measure of the periodic heat gain absorption capability of building elements. The deviation in indoor environmental temperature from the daily mean will be found from

Appendix 3 includes the calculation of this temperature swing for one timepoint, 15.00 h, which is likely to be the peak.

If we set a limit to the acceptable temperature swing (or deviation from the mean), the expression can be turned around to determine the required admittance. For lightweight elements the admittance is practically the same as the U-value. The admittance of heavy building elements depends partly on their thermal capacity (i.e., the product of their mass and the specific heat capacity of their material), but partly also on their conductivity and surface qualities. It is therefore a better measure to employ than the thermal capacity alone. Tabulated admittance data are available in many publications (e.g., CIBS Guide), but it can also be calculated by a fairly lengthy matrix operation.

This pattern-matching method is then a useful tool for the sketch-design stage. When the major design decisions have been made, any of the other available tools (e.g., the SLR method or some of the more sophisticated thermal response simulation programs) may be employed.

Passive Cooling

does not exceed the upper comfort limit.

The second law of thermodynamics recognises that heat can only flow from a warmer to a cooler body or region. Thus passive cooling may appear to be contravening the second law. There are actually only two processes which could be relied on:

1. dissipating heat when the ambient temperature is low and using a thermal storage effect to dumpthe surplus heat during overheated periods,

2. evaporative cooling.

The second of these is applicable in hot-dry climates and there it is very effective, both in passive and active form. The first one is also more effective in hot dry climates, where the diurnal temperature variations are larger. In warm humid climates the alternative is to have a completely open, fully cross-ventilated building, with no significant thermal mass. This could not achieve indoor temperatures below the ambient, but the thermal sensation of occupants may be reduced by a physiological cooling effect. This approach is well documented, it is the accepted solution for humid climates. The present discussion will be restricted therefore to the former approach, which relies on storage effects.

The two approaches are mutually exclusive, they would demand building solutions which are basically different, therefore a choice must be made very early in the design process.

greater than 10 K in the negative direction and the indoor temperature swing can be kept well within the ±2.5K suggested earlier. It is to be examined how far we can go with such systems in warm-humid climates.

With the traditional cross-ventilated system the maximum acceptable air velocity for domestic interiors would be about 1.5 m/s, which would cause a reduction of about 5K in temperature sensation. As mentioned earlier, psychological factors may favour the ventilated situation, so (with some degree of arbitrariness) we suggest that the closed building must be at least 7K cooler than the lightweight one to be acceptable.

Fig. 6 is the summary of an extensive study (Szokolay, 1977) carried out for Darwin, Northern Australia, which has a very humid and warm climate. Temperature profiles (produced by using the thermal response program (TEMPER) are given for a house with several different constructions for a January day, when the outdoor peak temperature is 32 °C and the minimum is 24°C (i.e., a diurnal range of 8 K). The conventional lightweight house goes up to 37°C. The heavyweight building only slightly exceeds 29°C and if night-time ventilation is introduced, the temperature is further lowered.

Fig. 6 Temperature profiles of a house with four different constructions in a warm-humid climate (Darwin, lat. −12.5°, January).

in this variant is only 2.2K, which is excellent, it would be keeping even the indoor peak within the comfort zone: the peak would be 7.9K lower than in the lightweight one, but only 2.8K lower than the outdoors, thus it would not persuade people to change over to this type and give up cross-ventilation.

Perhaps a better approach would be to improve the lightweight variant (reducing solar gains by better shading and roof insulation). This – as shown in to less than 1K and the peak to about 33°C, thus with the added physiological cooling effect it would produce about the same conditions as the heavy variant, whilst preserving the psychological advantage.

One could conclude that passive cooling in hot-dry climates by a combination of storage effect and evaporative devices is relatively easy. The indoor environment created would be more pleasant than the hot and dusty outside world, so the system would be psychologically also desirable. Not so in warm-humid climates, where the luscious vegetation outside would be more attractive than the stuffy and stale interior of a building, even if it were actually cooler than the outside. The conventional wisdom of the cross-ventilated building is confirmed. Should the breezes fail, a lowpowered ceiling fan would do the same job, with very little compromise of our passive principles.

VISUAL COMFORT

In discussing visual comfort we have to restrict ourselves to luminous comfort. The former term would include the objects of seeing (over which we have little or no control), whilst the latter concerns only the questions of lighting, i.e., the facilitation of the visual process. The former could be construed to include most, if not all, the subject area of architecture: buildings pleasing to look at, the control of views from these buildings, even the creation of pleasant views, interesting or even stimulating views.

Luminous comfort itself is a very broad subject. Human tolerance and adaptability to lighting conditions is far broader than to thermal ones. Thermal preferences may range between 18 and 30°C, at the extreme, but we are able to read a newspaper at the light of the full moon (about 0.1 lux), or in full sunlight (some 100 000 lux). Six orders of magnitude, or a factor of a million. Obviously, neither of these two extremes would be considered as comfortable, but luminous comfort is a much more ill-defined term than thermal comfort.

Some years ago I compared recommended illuminance levels for several tasks from various countries (Szokolay, 1980). Just to give one example: that of drawing board lighting:

The design target adopted will depend on socio-cultural and economic, rather than technical or scientific factors. Hopkinson’s (1963) visual efficiency graph (Fig. 7) clearly illustrates the law of diminishing returns. Illuminance is linearly proportionate to energy consumption, but gives logarithmically reducing benefits in terms of visual efficiency. We should probably aim at the upper-middle region of the graph, but the precise choice will be rather arbitrary.

Fig. 7 Visual efficiency as a function of illuminance (after Hopkinson, 1973).

Illuminance of the work-plane, important as it may be, is however not the only criterion of good lighting. Recent developments in lighting design are based on the consideration of the whole of the visual field, rather than the work-plane only.

NATURAL LIGHTING

Be it natural or artificial lighting, the likelihood of glare, the direction (vector) and directionality (vector/scalar ratio) of the lighting will be important aspects to consider. In electric lighting design these aspects can be controlled (even if they rarely are). In natural lighting this is much more difficult. For the present purposes however, this is our main concern.

Daylighting

In a room, particularly in a deep room, daylighting from vertical windows in one side-wall will cause a very low illuminance vector (see Fig. 8). Looking at a person between me and the window, I will see only a silhouette of the figure or the face. Several studies (e.g., Cuttle et al., 1967) have shown that most people, in looking at a human face, prefer a vector altitude of about 30", if the vector/scalar ratio is around 1.2 − 1.3, but much higher with larger vector/scalar ratios (e.g., 75° with a v/s ratio of 1.5, as shown in Fig. 9). (The v/s ratio varies between 0, in a fully diffuse field, with no directionality and 4, given by a single beam of light, with no diffuse component.)

Fig. 8 Illumination vectors in a side-lit room.

Fig. 9 Preferences for vector altitude versus vector/scalar ratio (after Cuttle et al., 1967).

So, if comfortable viewing conditions are to be created, the vector/scalar ratio is to be kept low (more diffuse light) and an attempt must be made to raise the vector altitude (in side-lit rooms there is little likelihood of overdoing either of these). The former aim can be approached by using light coloured room surfaces and both aims could be served by getting some light into the rear part of the room.

The introduction of PSALI (permanent supplementary artificial lighting of the interior) was triggered by this desire. It serves both purposes and it is a well-established technique. It also tops up the work plane illuminance at the rear of the room and it does all three jobs whilst maintaining the daylit character of the room. Compared to full artificial lighting it gives a substantial energy saving, but it still uses precious energy – so the purist would consider it as active control which should be avoided, if at all possible. It remains to be seen to what extent this would be possible.

The whole method of daylight design practised in the U.K. and northern Europe is based on the 15th %-ile outdoor illuminance values. We determine the fenestration to give a specified daylight factor, so that the design illuminance would be exceeded 85% of the time between sunrise and sunset. The target is set fairly low: the typical 2% daylight factor with the 5000 lux design sky illuminance value would only give 100 lux illuminance. The logic is that most of the time the illuminance will be higher, and when it becomes too high, then negative controls (curtains, blinds) can be used without too much difficulty.

Sunlighting

Under such overcast design sky conditions there is very little one can do to get some daylight into the rear part of the room and to improve the vector altitude and vector/scalar ratio, without using PSALI. Such skies would occur however very rarely in more pleasant climates.

In my own town, as Fig. 10 shows (from Ruck, 1982), 5000 lux illuminance is exceeded 92% of the time (8.00 − 17.00 h), and – on average- we have 7.5 h clear sunshine per day. It seems ludicrous to design our fenestration for that 8% of daytime hours. Beam sunlight is a very powerful tool. It can be directed by optical devices (mirrors or prisms) to reach the ceiling at the rear of the room, from where it would be diffused and would perform the same three functions as PSALI:

Fig. 10 Cumulative frequencies of outdoor illuminance in Brisbane, lat. −27.5° (after Ruck, 1982).

– adding to the illuminance of the work-plane,

– elevating the vector altitude,

– lowering the vector/scalar ratio.

Work on the development of actual reflector or refractor devices is in progress at several institutions (e.g., Lawrence Berkeley Laboratories and the Uni. of N.S.W.) and several, rather wild ideas have also been suggested. A group of my students proposed the device shown in Fig. 11, which would collect solar radiation by a parabolic dish, transport it through a bundle of optical fibres and diffuse it from a device not unlike an electric luminaire.(Architecture students are excellent generators of ideas, but rarely have the skills or perseverance to bring them to realisation.)

Fig. 11 Proposed light collector and distributor system (Uni. of Qld. fourth year Architecture students, 1979).

I have great expectations for the next decade – or so – for the increasingly widespread use of beam sunlighting. This would not only improve luminous comfort but would also lead to significant energy savings.

INTERACTION

Passive thermal and lighting controls are unavoidably interdependent, as the sun is our only source of both heat and light. In cool or cold climates some passive solar heating systems (e.g., direct gain) can create quite unpleasant and glary lighting conditions. Excessive use of Trombe walls have the opposite effect: creating rather dark and gloomy interiors. The more successful applications usually consist of a combination of two or more types of passive systems.

In warm climates daylight is the desirable external influence, whilst the thermal effect of solar radiation is undesirable. If – as in the suggested beam sunlighting methods – solar radiation is brought into the space, it will also have a thermal effect.

A quick comparison can be made between PSALI and beam sunlighting for the same task. The luminous efficacy of sunlight (for altitude angles greater than 25°) is about 117 lumens/watt. To get – say − 400 lux illuminance over a10 m² area we need a flux of 4000 lumens, so we must bring in 4000/117 = 34 W of solar radiation. The same lighting could be achieved by a 2.4 m long artificial daylight fluorescent lamp, rated at 125 W, which (including the ballast) would give a thermal load of 156 W. Thus in terms of heat load the electric lamp is some 4.6 times worse than sunlight.

The trouble with conventional side-lighting is that to get an acceptable level of daylight at the rear of the room, the lighting near the window will be excessive, with the accompanying large amount of solar gain. This solar gain could be drastically reduced if sunlight were to be directed only to where it is needed and only as much of it as needed. This would mean the use of some beam directing device, which has been assumed in the above comparison.

Without this we may have an unresolvable dichotomy: we employ a shading device to control solar heat gain and thereby we reduce daylighting so much that electric lighting will be used, which in turn will impose a significant thermal load on the building, be it passively or actively controlled.

REFERENCE

Arens, E., Zeren, L., Gonzalez, R., Berglund, L. and McNall, E. (1980). A new bioclimatic chart for environmental design. Intern. Conf. Bldg. Energy Management (ICBEM), Pavoa de Varzim, Portugal.

Auliciems, A. (1981). Global differences in indoor thermal requirements. ANZAAS conf., Brisbane.

Berglund, L. G. Revised standards on thermal conditions for human occupancy. Habitat International. 1980; 5(3/4):525–532.

Cuttle, C., Valentine, W.B., Lynes, J.A. and Burt, W. (1967). Beyond the working plane. C.I.E. Proceedings, pp. 471–482.

Fanger, P. O.Thermal comfort. Copenhagen: Danish Technical Press, 1970.

Hopkinson, R. G.Architectural physics: Lighting. London: HMSO, 1963.

Humphreys, M. A. Outdoor temperatures