Workshop on Passive Cooling

0 downloads 0 Views 14MB Size Report
Apr 4, 1990 - maintain an air temperature difference or protect from precipitation. ... heat for comfort by combustion is almost as universal as shelter ..... Their discussion is out of the aim of this paper, as much each ..... Such relations fit well with subjects in a climatic chamber for a few ...... loads without need for reheat.
INSTITUTE FOR SYSTEMS ENGINEERING AND INFORMATICS Non Nuclear Energies

WORKSHOP ON PASSIVE C001ING Ispra 2-4 April 1990

JOINT RESEARCH CENTRE EUR 13078 EN

COMMISSION OF THE EUROPEAN COMMUNITIES

5 INSTITUTE FOR SYSTEMS ENGINEERING AND INFORMATICS Non Nuclear Energies

WORKSHOP ON PASSIVE COOLING Ispra 2-4 April 1990

Organized by Joint Research Centre Directorate-General XII for Science Research and Development

Edited by E. Aranovitch Joint Research Centre E. de Oliveira Fernandes Universidade do Porto T.C. Steemers Directorate-General XII

PARL EURQP. BibÜoth. N.C./EUR

JiofS

CL

EUR 13078 EN

JOINT RESEARCH CENTRE COMMISSION OF THE EUROPEAN COMMUNITIES

1990

Published by the COMMISSION OF THE EUROPEAN COMMUNITIES Directorate-General Telecommunications, Information Industries and Innovation Bâtiment Jean Monnet LUXEMBOURG

LEGAL NOTICE Neither the Commission of the European Communities nor any person acting on behalf of the Commission is responsible for the use which might be made of the following information.

Cataloguing data can be found at the end of this publication.

Luxembourg: Office for Official Publications of the European Communities, 1990 ISBN 92-826-1690-8 Catalogue number: CD-NA-13078-EN-C © ECSC - EEC - EAEC, Brussels-Luxembourg, 1990 Printed in Italy

PREFACE

The development of biochmatic architecture is becoming a subject of increasing attention and interest the world over. This field not only has an enormous potential for energy savings and pollution reduction but is also an example where technology and architecture interact for the greater benefit of our society. In that context DG XII and JRC have organized together this Workshop on a specific topic "Passive Cooling" which has attracted to Ispra both architects and engineers. The subject is of course of major interest for climatic regions typical of the Mediterranean Basin where air-conditioning power, in summer, is increasing rapidly to meet with modern standards of comfort both in dwellings and working places. For instance in Greece the new-installed air conditioning power has increased 9-fold the last three years. The electricity load of the Italian electricity grid reached, for the first time, its yearly maximum in the summer of 1988 most likely because of power supplied to air conditioning equipment. Similar trends are observed in Spam and Portugal. An important fraction of the cooling load in the building sector could be saved by applying passive cooling techniques. However, it should be recognized that this field has not reached the same level of knowledge and maturity as solar passive heating because of a late start and the complexities due to interactions with the problems of human comfort. It has seemed therefore indicated to give now a higher priority in the European R & D programmes to "Passive Cooling" in an attempt to close that gap. The aims of this workshop have not only been to assess the state of the art but also to identify areas where future research is needed, to prepare actions which could contribute to prenormative and normative activities, to stimulate future collaborations. The organizers hope that the readers of these proceedings find that these objectives, at least in part, have been achieved.

CONTENTS

Agenda

7

List of Participants

9

SESSION 1

SESSION 2

SESSION 3

SESSION 4

SESSION 5

SESSION 6

Comfort

13

Comfort and Passive Cooling Nick Baker - The Martin Centre, University of Cambridge (UK)

15

Thermal Comfort X. Berger- C.N.R.S., Sophia Antipolis, Nice (F)

35

Micro-Climate

49

Passive Solar Cooling Design Jaime Lopez de Asiam - School of Architecture, Seville (E).

51

Microclimate Jaime Lopez de Asiam - School of Architecture, Seville (E).

55

Climatic Control for the Open Spaces of the 1992 World Fair Jame Lopez de Asiam et. al - Escuela Tecnica Superior de Architecture, Seville (E).

57

Numerical Modelling of Ambience Factors in Urban Environment J. P. Peneau - School of Architecture of Nantes (F)

63

Solar Control Techniques

73

Solar Control Techniques S Yannas - Architectural Association School of Architecture, London (UK).

75

Heat Attenuation Heat Attenuation as a Mean to Limit Cooling Load and I mprove Comfort M. Antinucci - Conphoebus s.c.r.l., Catania (I)

101

Heat Attenuation and the Thermal Mass Effect G.F. Papársenos - Greek Productivity Center, Athens (GR).

111

Ventilation and Air Circulation

121

Ventilativi Cooling: State of the Art Θ. Fleury - ENTPE-LASH, Vaulx en Velin (F)

123

Improvement of Summer Comfort by Use of Evaporative Cooling JR Millet-CSTB, Marne la Vallee Cedex 2 (F)

135

99

Natural Cooling Techniques

141

Natural Cooling Techniques M. Santamouns - PROTECHNA, Athens (GR)

143

Expert System for Passive Cooling Design of Commercial Buildings G S \estnn étal - CNR-IEREN, Palermo I

155

Natural Cooling Techniques S A varez Dom nguez - Escue a Superior de Ingenieros Industriales, Seville (E)

165

SESSION 7

Monitoring in Cooling Monitoring in Passive Cooling P. Wouters - Belgian Building Research Institute, Brussels (B) An Approach to Summer Performance Data Analysis in Full Scale Buildings J.A. Cusidó et al. - Universität Politècnica de Catalunya (E)

SESSION 8

SESSION 9

SESSION 11

SESSION 13

CONCLUSIONS

185 193

Performance Criteria and Integration Performance Criteria/Integration A. Rabl - Centre Energétique, Ecole des Mines, Paris (F)

207

195

209 221

Design Support and Tools Passive Cooling Design Support and Tools R. Serraetal. - School of Architecture, Barcelona (E)

233

Design Support and Tools for Passive Cooling: Perspectives for European Research J.R. Peckham - JRC, Ispra (I)

241

Model Development Computer Modeling and Cooling Strategies F. Butera - et al. - University of Palermo (I) A General Model for Cooling Design V. Calderaro - University "La Sapienza', Roma (I)

SESSION 12

177

Additional Heating and Cooling Auxiliary Environmental Control in Passively Cooled Buildings £ Maldonado et al. - University of Porto (P)

An Approach to Integrating Passive Cooling Devices in Buildings A. Frangoudakis - CRES, Koropi (GR)

SESSION 10

175

231

251 253 263

Design Support for Architects Design Support for Architects J.O. Lewis - University College, Dublin (El)

273

Miscellaneous Presentation of the ENEA/JRC Mediterranean Test Cell

289

291

Data Treatment of the ENEA-JRC Test Cell H. Bloem - ISEI - JRC, Ispra (I)

293

Reports of the Rapporteurs

299

275

AGENDA

Monday, 2 April 1 9 9 0 14.00

Welcome Address E. Aranovitch Introduction T.C. Steemers Presentation of the Workshop £ de Oliveira Fernandes Chairman

14.30

Session 1:

Comfort Ν. Baker X. Berger

15.20

Session 2:

Micro-climate J. Lopez de Asiain J. P. Pénau

16.30

Session 3:

Solar Control Techniques S. Yannas

17.00

Session 4:

Heat Attenuation M. Antinucci G.E. Papársenos A. Canha da Piedade

Tuesday, 3 April 1 9 9 0 09.00

Session 5:

Ventilation and Air Circulation B. Fl eury

09.30

Session 6:

Natural Cooling Techniques M. Santamouris G. Sil vestrini S. Alvarez

11.10

e S ssion 7:

Monitoring in Cooling P. Wouters J.A. Cusido

12 00

e S ssion e :

Additional Heating and Cooling E. Mal donado

14 00

e S ssion 9:

Performance Criteria/Integration A. Pabl A. Frangoudakis

14.50

Session 10:

Design Support & Tools R. Serra R. Peckham

16.10

Session 11:

Model Development F. Butera F. Conti V. Calderero

Wednesday, 4 April 1 9 9 0 09.00

Session 12:

Design Support for Architects J.O. Lewis

09.30

Session 13:

Miscellaneous ENEA - JRC Mediterranean Test Cell E. Aranovitch

10.30

Reports of the Rapporteurs J. Page C. Bofía P. Wouters

11.15

Panel - Audience Discussion Panel Members: £ de Oliveira Fernandes J. Page C. Bofía D. Kodonas E. Aranovitch T.C. Steemers

LIST OF PARTICIPANTS

S. Alvarez

Escuela Tecnica Ingenieros Industriales Universidad de Sevilla Avenida Rema Mercedes 41012 Sevilla-ESPANA

M. Antinucci

CONPHOEBUS Via G. Leopardi 148 95127 Catania-I TALI A

E. Aranovitch

Joint Research Centre Non Nuclear Energies Service T.P. 261 21020lspra(VA)-ITALIA

Ν. Baker

The Martin Centre for Architectural and Urban Studies University of Cambridge Department of Architecture 6 Chaucer Road Cambridge CB2 2EB - UK

X. Berger

Groupe d'Ecothermique du C.N.R.S. Université de Nice Av. Valrose 06034 Nice Cedex - FRANCE

H. Bloem

Joint Research Centre Non Nuclear Energies Service T.P. 450 21020 Ispra (VA)-ITALIA

C. Boffa

Istituto di Fisica Tecnica Politecnico di Tonno Corso degli Abruzzi 24 10129 Tonno-ITALIA

L. Bourdeau

C.S.T.B. B.P. 141 Sophia Antipolis 06561 Valbonne Cedex - FRANCE

V. Bouriotis

Center for Renewable Energy Sources 6 Frati str. Fousia 19400 Koropi-Attikis - GREECE

F.M. Butera

Universita di Palermo Dipartimento di Energetica Viale delle Scienze 90128 Palermo-I TALI A

V. Calderaro I nternational Solar Energy Society Sezione Italiana V a Bormida 2 00198 Roma- ITALIA

R. Colombo

Joint Research Centre Non Nuclear Energies Service T.P. 450 21020 Ispra (VA -I TALI A

F. Conti

Joint Research Centre Non Nuclear Energies Service T.P. 261 21020 Ispra (VA)-ITALIA

Puigdomenech

Escuela Técnica Superior d'Arquitectura del Valles P.O. Box 508 08220 Terrassa - ESPANA

G. De Giorgio

Politecnico di Milano Dip. Energetica P.zza L. da Vinci, 3 20133 Milano-I TALI A

E. de Oliveira Fernandes

Universidade do Porto Rua D. Manuel II 4003 Porto Codex - PORTUGAL

M. del Rosario Heras

Division Solar IER-CIEMAT Avda Complutense 22 28040 Madrid - ESPANA

Β. Fleury

ENTPE LASH Rue Maurice Audin 69518 Vaulx en Velin - FRANCE

*

P. Fragnito

Progettazione e Ricerca C.so Como, 9 20154 Milano-I TALI A

A. Frangoudakis

Center for Renewable Energy Sources 6 Frati str. Fousia 19400 - Koropi-Attikis - GREECE

J.O. Lewis

Energy Research Group School of Architecture University College Dublin Richfield Clonskeagh Dublin 14-I RELAND

J. Lopez de Asiain

Escuela Tecnica Superior de Arquitectura Universitad de Sevilla Avda de la Reina Mercedes 41012 Sevilla-ESPANA

E. Maldonado

Universidade do Porto Dept. de Engenharia Mecanica Rua dos Bragas 4099 Porto Cedex - PORTUGAL

J.P. Millet

C.S.T.B. B.P. 141 Sophia Antipolis 06561 Valbonne Cedex - FRANCE

J. Page

15, Brincliffe Gardens Sheffield S11 9BG - ENGLAND

G.F. Papersenos

13 Ergotimou Str. 11634-Athens-GREECE

R. Peckham

Joint Research Centre Genie Systèmes de Fiabilité

10

T.P. 321 21020 Ispra (VA) ITALIA

J.P. Pineau

Ecoe d'Architecture de Nantes Laboratoire CERMA Rue Massenet 44300 Nantes - FRANCE

A. Rabl

Centre d'Energétique Ecole des Mines de Pans 60, Boulevard Saint-Michel 75272 Pans cedex 06 - FRANCE

G. Rizzi

E.N.E.A. - sede di Ispra c o Joint Research Centre 21020 Ispra (VA) I- TAL IA

M. S antamouris

Protechna Ltd Thoemistokleous 87 10683 Athens-GREECE

H. Coch

Departament de Construccions Arquitectoniques I Universidad Politécnica de Catalunya Av. Diagonal 649 08028 Barcelona - ESPANA

T.C. S teemers

Commission of the European Communities DG XII - E Rue de la Loi 1049 Brussels-BELGI UM

P. Wouters

Belgian Building Research Institute Avenue Pierre Holoffe 21 1342 Limelette-BELGI UM

S. Yannas

Architectural Association school of Architecture 34-36 Bedford Square London WC1Β 3ES - ENGLAND

S. Zabot

Regione Lombardia Servizio Energia Via F. Filzi, 2 Milano - ITALIA

P. Zegers

Commission of the European Communities DG XII - E Rue de la Loi 1049-Brussel-BELGIUM

11

April 1990 workshop on Passive Cooling Joint Research Centre, Site of Ispra

12

Session 1 Comfort

Comfort and Passive Cooling N. Baker The Martin Centre - Department of Architecture - University of Cambridge UK

ABSTRACT. This paper is concerned with aspects of thermal comfort which are particularly relevant to passive cooling. A fundamental difference between cooling in a warm climate and heating in a cool climate is recognised due to the relationship between the physiological neutral temperature and ambient heat sinks. The need for a comfort performance assessment of a building design is established. The notion of person cooling as distinct from space cooling justifies the need for a behavioural comfort model as distinct from a fixed state model. Brief results from simulation and field studies are offered to support this view. This topic is identified as one needing further research. Other potential research topics include, psychological effects, airflow design for comfort, outdoor comfort, associated non-thermal comfort - glare and noise control. 1.0 INTRODUCTION Much work has been done in the field of thermal comfort. However a review of the literature reveals that in most of the work relating to buildings there is a pre-occupation with thermal comfort in predominantly underheated climates. Where overheating is studied it tends to bein situations, well above the limits of the normal comfort zone, where there are physiological effects of serious heat stress. In this paper we are concerned with thermal comfort within the upper limits of the comfort zone. Man in the natural unmodified environment would not survive for long. In a biological sense, the most important function of a building is to modify the environment in order to provide comfort. In this paper, we are concerned with the degree to which the building itself can accomplish that aim, or if energy consuming mechanical interventions are necessary, how the energy consumption can be minimised, fig 1. Man's pursuit of comfort is, of course, essential to survival of the species- it is not just whimpishness! It is a basic drive comparable with hunger and sex. So successful is man in response to this drive, that he has occupied a wider range of climatic zones than any other species on earth - and all without the aid of modern technology. Only the ultimate harshness of Antarctica required 20th century technology to enable man to survive. It is interesting to look at primitive shelters and to see what passive measures these illustrate. In fig 2 we see two simple shelters which are of geometrically similar form. One is to protect the nomadic herdsman of the Kalahari desert from the intense solar radiation, whilst the other protects the Indian of Tierra del Fuego, South America, from the deadly sub-zero winds. In both cases the shelter cannot be described as an envelope, it could not 15

maintain an air temperature difference or protect from precipitation. But in each case the structure modifies a single climatic parameter, choosing the one which is most uncomfortable, and ultimately life-threatening. As shelters become less primitive, so we see the range of environment functions increasing, and of course as pointed out by Rappaport (1) the influence of cultural and social factors, fig 3. But it can be argued that it was the growing sophistication of his shelter that moved man away from the edge of survival and released his energies for social and cultural development. Today, buildings still carry out that primary function of providing comfort. In doing so they consume nearly half of the total energy used, and with the current awareness of the global environmental effects of this, clearly the provision of comfort is imposing a heavy cost. It is this balance between comfort and energy cost to which we are addressing ourselves today. 1.2 Heating and Cooling It is a simple fact that a very small proportion of buildings in the world are actually cooled (in a thermo-dynamic sense), whereas a very large proportion of buildings are heated. Indeed, the non-passive provision of heat for comfort by combustion is almost as universal as shelter itself at primitive and traditional levels, fig 4, in all but the hottest climates. In contrast there'are few traditional ways of providing cooling, (as distinct from the prevention of overheating). Is there a fundamental difference then, in the case for heating and the case for cooling? The mean monthly temperature in Athens, for July, is about 28° C. By accepted comfort theory, for a person at rest dressed in 0.3 Clo, the neutral temperature is also 28°C. We realise that discomfort will exist for a number of reasons - the actual temperature will vary about the daily and monthly mean, and in a building, the temperature may be elevated above prevailing ambient conditions due to gains from solar or other sources. However in almost all cases there will be ambient heat sinks available, and thus the discomfort problem reduces to one of the heat being in the wrong place at the wrong time, over a relatively short time scale. This contrasts with the case for heating. The mean January temperature in Hamburg is 0°C. Even with a clothing level of 1.5 Clo a person at rest has a neutral temperature of about 19°C. Of ambient heat sources available, even the longest term, i.e. the ground, will be at about 9°C. No other passive thermal source is universaly available with the exception of solar energy, unless we turn to fossil sources. Thus there does seem to be a basic difference, and this relates to the human temperatures for survival in relation to climatic temperatures - i.e. in most cases prevailing temperatures, at least on a daily average basis, are far further below our neutral temperatures than above. This seems like good news for passive solutions to cooling. 16

Another difference relates to magnitude of the comfort zone. Fig 5 shows the results of s study carried out by Humphreys (2). It shows clearly that for a given activity, the tolerance band is greater for people at a higher temperature and lower Clo value than vice versa. However on the negative side, most human activities in buildings, including humans themselves, are heat producing. In an underheating situation these gains are assisting in providing comfort, whereas in an overheated situation they are adding to the problem. Casual "coolth" gains are extremely rare. Furthermore, the high grade energy of solar radiation can be controlled to give substantial useful gains in the heating situation, whereas in cooling it is more often a cause of the problem than a contribution to a solution.

1.3 The prevention of overheating and the provision of cooling This brings us to this useful distinction. Where ambient conditions are such that in the absence of direct solar radiation, the upper comfort limit is not exceeded, our main concern is to prevent the building from providing an environment which is worse than that outside. We can show that for much of Europe this is the prevailing climatic condition, and even when it is exceeded, it is for a relatively short time. This assertion is supported by the traditional use of the outdoors for many domestic and social activities , in Southern Europe - and it is of course, the "holiday environment". It is only in modern working situations that conventions are forcing us indoors. Thus the problem becomes one of prevention of overheating, ie the minimising of gains from solar radiation and from internal sources, rather than the provision of coolth sources. It suggests that this should be our first line of defence - e.g. we should be concerned with shading devices long before solar chimneys. There are some cases where the climatic situation and the building use may demand that internal conditions are significantly cooler than ambient, or where internal gains are by necessity very high. Then the provision of cooling applies. Conventionally this is provided by mechanical refrigeration, and less commonly, but of special interest to us here, by passive cooling methods, fig 6. I think that this distinction is important. It is analogous to the earlier days of solar heating where much confused thinking seemed to prevail about the relationship between conservation and solar heating. Arguments raged about solar fraction, solar contribution, absolute energy consumption etc. Now a more holistic view is taken and few would argue that to meet the objective of maximising comfort for a minimum energy consumption, conservation is the first step. (For convenience, we shall still referto both prevention of overheating and the provision of cooling by the inclusive term passive cooling.) 17

2.0 THE ASSESSMENT OF COMFORT PERFORMANCE 2.1

Reduction and Avoidance

We now have to approach the problem of how we assess passive cooling. First however we need to consider two possible outcomes of adopting passive measures - avoidance of mechanical cooling, and the reduction of cooling loads, fig 7. If our passive measures, predominantly prevention of overheating, lead to a reduction of air-conditioning costs, then performance assessment is straight forward, being analogous to displaced auxiliary heating in a passive solar heated building. But supposing our objective is to avoid mechanical cooling altogether, then we do not have an energy consumption upon which to base our assessment. Clearly the answer is a comfort performance index. We need a way of assessing a design proposal in such a way to show that the comfort criteria are met and thus the avoidance strategy can be pursued. Before going on to discuss the nature of this comfort performance index I would like to stress the importance of the avoidance option. It is clear that without a mechanical cooling system, no cooling energy can be used. It is not so clear how much energy will be used if a mechanical cooling system is present. If we knew that it would only be used in extreme conditions, then provided the design used passive measures to reduce the cooling load, the energy consumption might be very small. But experience shows that once having accepted artificial conditioning, occupants will begin to adjust their expectations to require lower temperatures than they would accept in a nonair-conditioned building. Indeed they will begin to "dress for the environment" - in the case of the work place for example, it could be argued that in summer air-conditioning is provided in order that the male occupants can still comfortably wear the business suit! Certainly our experience in the UK in winter, supports this view. It is common in both domestic and non-domestic buildings to find higher indoor temperatures in the winter when the heating systems are under automatic control, than in the summer, when it is common practice to switch off heating systems altogether. This reverse acclimatization is supported by the trend to wear very light clothing indoors in winter. There is also evidence that occupants of air-conditioned buildings where they have little control over their environment, are far more critical in their thermal requirements. Thus if mechanical cooling were present in a building, and particularly it was indistinguishable from the system of heating, it seems likely that people would develop a demand for cooling to temperatures considerably below that which they would be happy in a passive building. 2.2 Comfort Performance Index We have already established the need for a comfort performance assessment of a proposed building. Fig 8 shows schematically an approach where the comfort performance of a proposed building can be compared 18

with a base case. The purpose of the base case is to establish reasonable expectations for performance in the particular climate and for the particular building type. This would allow any improvement (or worsening) of performance to be measured against a value and the feed back to be used to optimise the design. Once the comfort performance criterion is met, the air-con equivalent can be used for the purposes of economic assessment. Here, energy consumption values for typical buildings of the same use type and location will be used to establish the value of the energy saved by taking the avoidance option. The comfort performance index must be based upon a subjective comfort index, as distinct from a single objective parameter such as air temperature. Comfort models which take account of the environmental parameters air temperature, radiant temperature, humidity and air movement, are well established as typified by the work of Fanger (3), and described in detail in the next paper. By considering heat loss to the environment by the four mechanisms of convection, radiation, evaporation and conduction, a single equivalent or operative temperature can be derived for any set of environmental parameters. When this prevailing temperature is compared with a neutral temperature, at which metabolic gains and environmental losses are in balance, the discrepancy can be related numerically to a subjective response. In the classic work by Fanger these responses, referred to as the Predicted Mean Vote (PMV) are ascribed a value viz +3 +2 +1 0 -1 -2 -3

hot warm slightly warm neutral slightly cool cool cold

Whilst this predicts a subjective response to an environmental state, it does not indicate the subjects satisfaction with that state. Clearly if we applied this to our proposed index, we would only be able to say for example "that the building would be warm for 60% and hot for 20% of the time". Fanger extended the PMV to include a statistical indication of satisfaction, or rather dissatisfaction with the Percentage Persons Dissatisfied (PPD), which derived from a direct relationship with PMV as indicated in fig 9. Thus we have at our disposal, a well proven comfort index. But there are problems. In fig 10 we can see that PPD values are not only dependent upon environmental parameters, but also the human parameters of activity level and clothing. What values of these will we assume in evaluating our building?. Furthermore, work of Humphries and others suggests that we should not be using a neutral temperature derived from heat balance, but rather one which reflects a degree of acclimatisation. There may also be psychological factors which influence the neutral temperature, or rather more subtly, the shape of the PPD curve as we move away from zero. When applying such an analysis to a heated building, these uncertainties may result in uncertainty in where to set the thermostat. In the case of testing a building design where we are trying to decide between the passive and 19

mechanical options for cooling, uncertainty will have more serious consequences. 2.3 Variation of Comfort with Space and Time A greater uncertainty is due to the fact that comfort conditions are neither constant in space, nor in time. Of the four environmental parameters influencing comfort, air temperature, radiant temperature, air movement, and relative humidity, the first three of these are likely to vary significantly from room to room, and within a room. They will also vary with time, and it is unlikely that any passive building will experience steady conditions throughout the occupied period. Fig 11 shows the results of simulations carried out by Newsham (4) at the Martin Centre in Cambridge. The simulation model divides the room into 27 cells and calculates radiant and convective exchanges between the room air and 54 surface nodes. Temperature gradients, vertical and horizontal, are predicted by a simplified model, but airflow itself is not modelled. The results shown here give map out the variation in space and time of the PPD, calculated from radiant and air temperature alone, for the room in a predominantly overheated situation. The climatic data is for a day in July, in the UK, and the rather low ventilation rates have led to overheating. The non-uniformity in space and time is very evident. One of the investigations of the study was to see if when an occupant moved around the room, seeking the most comfortable point, this would lead to a significant reduction in overheating (or underheating), as indicated by hours of PPD > 20%. Fig 12 shows the path taken by the "comfort seeking occupant". The results on the hours of discomfort were dramatic - the annual overheated hours were reduced from 570 hours to 130 hours. We would not like to suggest that it is always desirable that occupants have to seek out a comfortable position. It is not always practicable -in a densely occupied office or classroom for example. However it is probably quite close to actual behaviour, especially in houses where the occupant ususally has a freer choice of position. In cool climates, the traditional approach to heating, i.e. the fireplace or stove, led to considerable temperature gradients (both air and radiant), and normal occupant behaviour would include moving around until comfortable. And indeed, the need for this may be regarded as a positive attribute of the room. In a field study by Griffiths (5), PMV values calculated at a single point, show very poor correlation with actual subjective replies. The reasons for this are not quite clear from the study, but the fact that the correlation was best for the offices, where the was a greater probability of uniformity and less opportunity for occupant choice of position, suggests that spatial distribution may have had something to do with it. The results also showed that in a space tending to underheating, people are more tolerant of the recorded low temperatures, than expected, suggesting that the non-uniformity may have been used to provide local comfort. It seems that a combination of common sense and the rather scant evidence above, suggests that a comfort performance index must take account of 20

variation in time and space of predicted comfort conditions, and must take account of how the occupant is likely to behave. This behavioural aspect should also include clothing and activity level, and possibly other unspecified (so far) factors. It must also include the operation of building controls such as blinds and opening windows. 2.4 Behavioural Models This all points to an interest approach differing from the conventional way of assessing building performance. Hitherto, passive building performance assessment has concentrated upon the energy performance in heating situations. For this purpose, the occupant model has been very, very crude usually the occupant is modelled as an inert temperature sensor located in the middle of the room. Has this mattered, and if not, why not? Probably it has not mattered too much, because the most significant error would be in heating energy input, and the same systematic error would be made when making comparisons. Validations of precision models almost always involve data from test cells or simulated occupancy houses, where these behavioural aspects do not appear. However, where we are testing a building to see if the avoidance of mechanical cooling is justified, it is crucial that the influence of behaviour on comfort performance, is taken into account. Thus any assessment will require both a building model, to predict the effect of the building upon the climatic boundary conditions, and a behavioural model, to predict the response of the occupant. There are precedents for behavioural models in building performance assessment. A good example is in light-switching algorithms. These predict the probable response of an occupant to ambient daylight levels in a room, in order to determine the saving of electrical energy. It is interesting to note too, that in some respects the human response has proved more energy efficient than an automatic "objective" response. For example, an automatic switch-on with a datum of 300 lux for an office building has proved to be pessimistic since often people will tolerate far lower levels at the beginning and end of the day, provided they associate it with daylight and not insufficient artificial lighting. 2.5 Design Tools Any assessment model has the potential of becoming a design tool, provided it can respond to design parameters available early in the design development. The time and space dependency required in the behavioural model here, suggest that this would only be available from deep-analysis simulation. This requires much building data, apart from the so-far unknown spatial data required for the behavioural part of the model. At the Martin Centre we are currently working on a behavioural model which aims to quantify an occupants exposure to daylight. This incorporates a statistical position occupancy model, which basically gives the probability that a person will be in a specific place at a specific time. Rather than a simulation model, which due to the complexity of movements and occupancy 21

patterns of an actual building, we are trying to develop a parameterised model, using parameters such as depth from outside wall, distance from workstation, with special space-use weightings for circulation, rest rooms, toilets, etc. If such parameterised simplified behavioural models appropriate to thermal comfort can be derived, then it might be possible to combine these with simplified thermal models and produce comfort performance design tools. 3.0 In the second part of the paper a number of other aspects of comfort which have particular bearing on passive cooling, are now discussed. 3.1 Interior Environment Design The occupant centred comfort models discussed above, suggests that where conditioning of some kind, whether mechanical or passive, has to be provided, it would be most efficient to provide it on a local scale. This is already accepted, indeed preferred practice for lighting - i.e. task lighting, and as already discussed, until the concept of space heating was developed, was accepted (and perhaps preferred) practice for heating. Thus we must develop the concept of "coolth emitters". An opening in the external wall which can modulate and direct the prevailing wind onto the occupant, could be considered to be a passive coolth emitter. An electrical fan, operating on the same environmental parameter, could be regarded as the auxiliary equivalent. A cooled floor, cooled by a passive source such as a stream, or actively by chilled water, may also have advantages over the conventional convective volumetric cooling, fig 13. In this system, stratification and stability of the cooled lower layer of air provides an occupied zone, which in the case of the seated occupant is considerably less in volume than the whole room. Even furniture could be cooled, again passively or actively, and advantage taken of conductive losses from the body. This brings us on to the whole area of interior design. Not only should this be directed to solving the environmental performance in an integrated way, i.e. the interaction between requirements of daylight, airflow, view, shading etc, but also the psychological aspects must be considered. The substantive research is yet to be done, but common experience hints to us that our physiological expectations may be influenced by visual and other physical cues. I have long been interested in what I have called the "switched-off-escalator effect". We probably all have felt a curious sensation when walking down or up an immobilised escalator on the metro. Why is this feeling actually physical? We have only received visual cues to tell us that normally it is an escalator - we do not get the same feeling when we walk down an ordinary staircase. If the visual cues that tell us that we are walking down an escalator, effect 22

our muscles, posture and balance to such an extent, then maybe visual cues such as furnishing, finishes, view, planting etc can influence our expectations of thermal comfort. There has been some work in this area, mainly for the heating situation, and it did not show a positive effect. However this was investigating only the effect of colour. Maybe it is a "cocktail" of stimuli which have an effect - perhaps greater than the sum of the parts. If there is any validity in this suggestion, then it will promote a most exciting fusion of functional and aesthetic aspects of interior design. 3.2 Outdoor Thermal Comfort The discussion above relates in part to the use of outdoor spaces in warm climates. The traditional precedent is very well established, but it tends to be limited to recreation and relaxation. The design of the outdoor space, shows evolutionary development in some cases, e.g. patios and courtyards, but is often ad hoc, e.g. the side of the street. The use of an "outdoor room" for office or institutional buildings is rarely in the brief of an architect. But outdoor architecture really could have something to offer as a means of passive cooling in the broadest sense. Comfort analysis has been carried out on outdoor spaces in warm climates. Lauritano et al (6) produced Standard Effective Temperature predictions for Trapani, Italy, for shaded and unshaded situations. He concludes that there is a need to develop the model further to take account of the complex geometric geometry of the typical urban environment. Fig 15 shows a section of the Faculty of Humanities of the University of Seville. In extreme conditions, in order to maintain cooler than ambient air at the base of the atrium, cross-ventilation through the library would have to be curtailed. But if the library were occupied, ventilation would have to take place. The answer was to design the outside space to the north of the library as an outside room or study area, still maintaining the necessary book security, and providing a pleasant natural working environment at no energy cost for lighting or cooling. 3.3 Non-thermal comfort Finally, it must be noted that there are other aspects of comfort relating to passive cooling which are non-thermal, but may be interactive through psychological mechanisms with thermal comfort. Glare is a good example. It appears from traditional solutions that glare may be more acceptable in cool climates, than in warm climates. This must be due to association with thermal discomfort. However it means that shading design should consider the brightness and positions of insolated surfaces which can be seen by the occupants, as well as considering the role of shading to prevent direct radiation from entering the room or actually falling on the occupants. These constraints often conflict with the need to maintain sufficient daylighting levels, and make the design of shading devices no small problem. 23

Another non-thermal comfort problem, which may or may not interact psychologically, is that of noise control. Any naturally ventilated building is going to be more vulnerable to external noise, due to the openings required for ventilation. Also noise generated within buildings where rooms are interconnected to provide through routes for natural ventilation, may present problems. Undoubtedly, technology can find a solution to these problems. 4.0 RESEARCH TOPICS To summarise, I have identified a number of topics in human comfort which are of special relevance to passive cooling, which I believe are in need of research. If I have omitted reference to work which already answers these questions, then I apologise to the authors, but it is clearly good news that the work has been done. However, even when in detail the research area has been covered, there may yet be the need to interpret and apply the findings to the specific context of passive cooling. 1.0

BEHAVIOURAL COMFORT PERFORMANCE MODELS Sub-tasks 1.1

Occupant studies (i) dress, posture, activity, use of building controls (ii) psychological effect, comfort expectations, interior design

1.2

Building studies (i) 3D room models, airflow and radiation (ii) interior design, furnishings, layout etc (¡ii) "coolth emitters" and controls

2.0

OUTDOOR COMFORT DESIGN

3.0

NON-THERMAL COMFORT

3.1 3.2

Glare Noise

Figi The building acts as a moderator between the b io log ica I needs of the occupant and the climate

24

·» a.··'·.?*' ·

írs**"·'· '"#*

JL' » T . ^ * * * ­ . . "

. i ·

*^

­ ■■■; ■' ■ ■ * i ­ « ­ "" « * * ·

Sunbreak of the Kalahari n o m a d , Africa

Windbreak in Tierra del Fuego, S America

Fig 2. Two similar forms, the sunbreak and the windbreak each protect against the most threatening environmental stress

25

Pueblo Indian village, Central America

Indonesian house

W i n d catchers of Hyderabad, Pakistan

Fig 3. Traditional solutions where more sophisticated environmental responses permit greater cultural influences 26

Drawing of an early seventeenth­century cottage rebuilt by the Central Electricity Generating B oard at Coleshill and now part of a Field and Local Studi« Centre; an example of the transitional phase between a cen­ tral open hearth and a closed chimney breast. The hearth occupici one half of a small end bay with a timber­frame wattle and daub 'chimney' or canopy above. In other houses similar in plan but perhaps rather earlier in date, the narrow end bay might have been a smoke bay without any canopy or chimney with the stairs to the upper floor within this open smoke bay.

Fig 4. The use of fire to provide warmth is almost universal to man in cool climates.

Resting I Sitting \ Standir

J Walking

40

60 80 100 120 2 M. metabolic rate . W / m

140

160

Comfort zones for air velocity ι·ο m/s Fig 5. Comfort is attained for a greater range of activity levels for low clothing values and high temperatures. Source Humphreys ( ) 27

Provision of Comfort

PROVISION OF COOLING

PREVENTION OF OVERHEATING shading insulation high reflectance ventilation min. int. gains

PASSIVE

V

ACTIVE

V

V

Fig 6. The provision of comfort - first strategy.

Provision of Comfort

AVOIDANC;EOF AIR-COND ITIO Ν ING

COM FOI3T PERFORMANCE

Fig 7 The provision of comfort-second strategy

28

REDUCTIC)NOF COOLING LOADS

ENERGY SAVING

Climate

Base case

PPD

Design proposal

Air­con equiv

FIMUfílÃKlELl

PIBIFiSMAHlEO

PPD

INDEX of PERF.

Annual Cooling Energy

INDEX of PERF.

X

COMPARISON

feedback

Fig 8. An approach to cooling performance assessment

% 80 LÜ

fin

O
l ...). 2.3. Air temperature This well­known index is not without problem of valuation: expressions of thermal exchanges have to consider local skin and air temperatures, and air flows (laminar at the b o t t o m , transitional for trunk, turbulent for head). Ave­ rage values (as the four point system by Ramanathan for mean skin tempera­ ture) and formulas are adopted. 2Λ. Mean radiant temperature It is an average of the surrounding surface temperature weighted by surface area, ability to emit heat (radiosities), and placement (solid angles of view from any point P). Measurement is insured by using a globe thermometer (black spherical copper shell with a thermal sensor in the center). One problem is on the large temperature fluctuations existing in space and t i m e , mainly due to solar radiation: the large asymmetries make the felt c o m f o r t quite different from the one in a uniform environment. 2.5. Air velocity This parameter affects both convective and evaporative losses. Its control can be immediate and local. So, air velocity can often be pointed out to act against the fatal effects of w a r m t h , humidity, a c t i v i t y or radiation. Sensiti­ veness to air velocity is great: allowance can be made for the secondary conse­ quences attached to it (noise, smell, dust), and for the agreement felt by the kind of air agitation (draught, air renewal). As it's the relative movement air­body which is concerned, the air velocity Vair to enter in the relations is: Vair V * 0.30 (Act ­ 1) if Act > 1 Finally, due to permanent natural air 0.1 m/s air velocity always exists.

and Vair = V if not. movements inside a room, a minimum

2.6. Humidity Humidity only acts on latent respiratory and cutaneous exchanges. When swea­ ting is not required, these exchanges represent no more than 20% of the whole. In this case, humidity has a small impact on c o m f o r t , the air evaporating power (partial pressure of water vapour inside the air, compared w i t h saturated pressure of water vapour of the expired air, or at skin level) being great. Otherwize, variations by a few percents do not a f f e c t much evaporating power

37

and comfort so long as saturation is not brought near. The problems are rather Jinked to personal (breathing...) or inertia! masses (condensations) affections. So, it's by band that this parameter has to be considered. 3. PSYCHOLOGICAL ASPECTS Defining comfort as the sensation of well­being relates to a mental appre­ ciation: the "personage" aspect of man has to surpass the "thermal machine" one, except in health centers. Agreement, preference, decision are relevant from it, as much the thermal lived context corresponds with situations of adaptation (and still much more with situations of saturation or suffering). Climatisation are such situations where all the various thermal sensors are stimulated, acting for a local and global protection of man. This protection can be ensured either by search for a null balance, or simply by search for a possible action of any particular system which is in limit situation (sweating in saturation...), even if neutrality has to be forsaken. It must be considered the declared comfort, the perceived comfort, the indifferent comfort. The first is linked to a given opinion, the second to an unconcious reaction, the third to an absence of appreciation (which does not necessarily corresponds with an agreement). Man is complex, and multiple appreciations can be simultaneously emitted, relative to a general well­being and local discomforts: for instance, pleasure of breathing a cool air, and disa­ greement of a too lighted environment. As it is in politics, the comfort appre­ ciation, and the decision for any action, is often relevant from a choice among several criteria, and so, from each one. Brain and thermal states supplement the five senses in our appre­ ciation of any environment. The judgement of one of them is conditioned by the state of the others, when it is extreme: for instance, spicy food, or predo­ m i n a t e red walls shift the thermal judgement to the warm. B esides this inter­ actions which are to be taken into consideration, the concept of comfort arises from various valuations. Due to differences between people, it is not possible to satisfy everyone at the same time. Looking for the highest possible percen­ tage of "comfortable" opinion needs establishing laws about the most influencing parameters in any given thermal context. These laws come from analysis of questionnaires, and correlate some physical variables to the feelings expressed (or lack of expression) by the way of a "predicted mean vote". 3.1. The Predicted Mean Vote PMV This index is commonly used to predict the average value of the subjectives responses of a group. A semantic scale from ­3 (cold) to +3 (warm) is used to express the thermal sensations. A correlation has been established with the comfort feeling: the range from ­1 (slightly cool) to +1 (tedious) corresponds with comfort. A correlation has also been established with the Predicted Per­ centage of Dissatisfied PPD (fig.2). % ­1

1 1 | [ 1 I |

1 1 1 1 1 1 1 1 1

1 1 1 1 1 | 1 |

1 1 1 1 I | 1 1 | 1 1 | 1 |

art

80· 60. 60·

..

^S

s^

40.

40 30.­

­

y'

­ ­ \­

==

\ ­ = ­ == = = = b , ­ ­

­

s

­ :

­

====­ ­

­ ==

­

y/ -

­= ­== ­­

= =?i

= == = = =

­== ­

=ΞΞ = ΞΞΞΞΞΞΞΞΜΞΞΞΞΞΞΞΞΞΞΞΞΞΞΞ=2ΞΞΞΞΞΞΞ|ΞΞΪ

a, Figure 2 Predicted Percentage of Dissatisfied, as a function of the Predicted Mean Vote (4­9) 38

20

"

\r ^r­

~X-

Σ

­ \

,ο 8 Κ.

!

~

¿

__X5

­

-i Ζ

­

y

J. Ζ __ +

5

­,Ζ ¿



±­'

ϊ

­

­

­

±

. . . . » ­ _|. _^ _μ ­^ 5 4Í­L­LLLJ—MII Ι Μ I J—LL_I_LJ 1I I 1, 1111 M I I -23 -1,5 -10 -0.5 0 05 10 1,5 20

P . M. V.

3.2. Perception of warmth Thermal n e u t r a l i t y , defined as the conditions in which a person desires neither a warmer, nor a cooler environment, is no more than a possible definition of c o m f o r t . Correlating perception of warmth and c o m f o r t is a usualy way adopted by scientists. The expected answer to question "how do you feel the ambient air ?" is given by a relation (10) such as: PMV - 0.251 Tamb + 0.013 RH - 7.2kk - 3.0 (Vair - 0.2) where Tamb is the temperature of a uniform ambience Tair = Trad, RH the humidity in percent. This relation convinces for sedentary a c t i v i t y , temperatures from 7k to 30 °C. Rohles and Nevins (11-12-13) proposed a similar relation (in addition +2.0 (Icio - 0.6), lower influence to air velocity). Fanger proposed a relation with a c t i v i t y level and the energetic balance as parameters (9). Such relations f i t well with subjects in a c l i m a t i c chamber for a few hours. They translate in terms of warmth perception the influences of parameters: + 0.1 m/s - * ► ­1.2 °C + 0.1 d o ­ * ► +0.7Ί °C (Rohles and Nevins) + 30°o RH ­ * ► +1.5 °C They need to be improved and checked for other thermal and environmental conditions. B ut, contrarily to these relations, the perception of warmth is not symétrie below and above a wished temperature (fig.3): 3 degrees below are felt slightly cool, in opposition with k above felt slightly w a r m .

( very hot \ [ hot

unpleasant

• · · • · ·

slightly warm

Figure 3 Warmth perception below and above a wished t e m ­ perature (in the present case 25.6 °C). The area of every c i r c l e is propor­ tional to the number of answers. The regression line points out a sensiti­ vity greater in the cold than in the warm (ik)

·



' ·

.

· ·



• ·%*M* · • ι y* \ · · · •

·

· * J » ^ ^

. . *

neutral

unpleasant

· · · · · .



·

*

·

i

1

d o 4r 0 6



act : sed.



20

30

25

22.9 2 7

25.6 '

29 5

,

35 'c

39

3.3 The mean wished temperature People tend to dress for the season and outside weather conditions. A survey by Humphrey (15) has related outdoor temperatures Tou with c o m f o r t indoors Tco. The prediction proposed (free running buildings) has a standard error of 1.0 °C and applies to the range 10 ­ 3 ^rrTC^.V-J.Æ, -WA«. " ....IJ,^¡111

Vegetation cooling effect — Shadow — Moisture — Dust absorption — Oxygène production

ν·;·ΙοΠοη Τ«. 3 5 - C Η· 4 0 * .

Tmod. Î2-C Η· 6 0 %

Woter ·. open circuit evaporotion 1

Fig.

■!./..

1. Vegetation cooling effect.

Fig.

A t «wporertoA point

2. Water cooling effect.

·' 'X-*

THERMAL CRITERIA

SELECT REFERENCE (UNOBSTRUCTED) APERTURE SIZE

SITE OBSTRUCTIONS TREES, VEGETATION

γ ««

^^^EMLUDÈ^­««*^^ < ^ DIRECT RADIATION J > ^ \ A L L YEAR^"""'^

N

»»

REVIEW: ORIENTATION TILT APERTURE AREA (LT CURVES OR HEATING CALCS)

BALANCE POINT TEMP FOR SHADING

► ADJUST ORIENTATION, TILT APERTURE AREA

t

Ν

SOLAR CONTROL REQUIRED?

ν

*

SELECT SOLAR CON-rROL TECHNIQUES (S) FOR ORIENTATIONS AND TILTS CONSIDERED

i ί

^φ­ OTHER DESIGN CONSIDERATIONS? ­ ^ ­ ASSESS OVERALL PERFORMANCE

J J 1

FIG. 7 Solar Control Design Procedure

85

T Q n , in conjunction with a threshold daylight datum, are also relevant parameters for controlling the action of adjustable blinds and louvres, manually or automatically, and may perhaps also apply as switching criteria for variable transmission glazing systems. Estimates for a wide range of values for the above parameters indicate that mean values of T b could range from below 10 °C to over 25 °C (for values of Τ ο}) between 24 and 29 °C). At the lower extreme of Τ^ ρ , spaces sited in the south of Europe will require almost all year exclusion of solar radiation under prevailing outdoor temperature patterns, while at the higher extreme most spaces in northern European regions will scarcely require any solar control. The balance point temperature can thus be a criterion in deciding design priorities. Clearly, the thermal capacity in the building structure will have a moderating influence, and depending on different values of thermal capacity the actual balance point at any given moment may be higher than the mean value. Parametric studies using dynamic simulation or the empirical concept of thermal admittance can be used to allow for this, and threshold temperatures for different space functions and values of the above parameters can be tabulated for use at pre-design stage. The LT curves [9] could be used to determine starting values for glazing areas. For spaces where solar heat gain is permanently undesirable the temperature criterion is still useful as an indication of overheating threshold but not a sufficient design criterion for sizing the aperture of openings. Daylight factors or other illumination targets can be used for sizing. Where shading devices are also to be applied, the area of opening could be adjusted to allow for the target illumination values. Baker [10] has suggested transmission factors for diffuse radiation with respect to louvres which can be related to target Daylight F actors. Figure 7 is an attempt to systematise the operations that need to be encompassed by a design procedure aimed at defining overall priorities as well as providing design support on solar.control. Most current software developed for the purpose of shading assessment do little more than emulate an heliodon. F or design support there is considerable scope to combine shadow and sunpatch prediction with daylight and thermal analysis and the possibility of accessing typological inventories of solar control techniques.

3.0 REVIEW OF SOLAR CONTROL TECHNIQUES 3.1 Solar Control By Geometry and Aperture Regulation Scope: Choice of orientation, tilt and aperture of surfaces as means of: - optimising annual influx of solar radiation; or, - minimising the incidence of global radiation at certain periods; or, - affecting the ratio between the direct and diffuse components of the irradiance reaching a surface.

^v \

\c

I 1

\

\

**^ \ '-'--

ORIENTATION

\

FIG. 8 Orientation, tilt, aperture 86

APERTURE SI ZE

Means of Solar Control: - Option of southern orientation as optimum where solar heat gain is required at times; option of northern orientation as optimum where solar heat gain not required. - Option of downward tilt of aperture (outward from building interior) to increase exclusion of irradiance; - Optimum" aperture size (e.g. using the LT curves as a starting point) on the criterion of a minimum total annual energy cost (heating, cooling and lighting, or cooling and lighting only; for non-airconditioned buildings indoor overheating threshold as upper comfort limit criterion). - Concept of a variable aperture, with minimum area based on daylighting criteria and maximum on solar heat gain considerations. Regional Variations and Topics for Study: The considerable variations between the north and south of Europe with respect to geographic latitude, outdoor temperature, and the magnitudes of incident solar radiation suggest the likelihood of significant differences in optimum values and appropriate design strategies for different building types. These differences have not been clearly identified in the literature so far, while many recent building projects across Europe appear to reflect homogeneity rather than differences in their "passive" design characteristics. There is scope for investigating the relative importance of orientation and aperture size, and to a lesser extent tilt, for southern European locations, both as means of solar control and for the overall environmental performance of buildings. Results would also provide a basis for comparing and assessing other means of solar control. Some hypotheses that appear to deserve study are given below: - The high gains to loss ratio that characterises south-facing glass in the south of Europe is often wasted by the recent trend in combining large glazing areas with low U-values for the opaque envelope, thus creating conditions of low solar utilisation and high risk, of overheating. In principle, there could be a number of trade-offs between orientation, aperture area, and thermal properties; it may well be that improvement of thermal properties is more relevant for glazing than for opaque envelope. Such trade-offs are likely to vary according to building type, as well as between mechanically conditioned and free-running buildings, and this is also an important area for study. The notion of variable aperture may be envisaged as using components combining the functions of solar/daylight control with night insulation (and security), the latter being also applied selectively during daytime to partly contract the area of exposed glazing. This is equivalent to traditional use of shutters in southern Europe. - North facing glazing has been anathema in recent years and this is clearly justified by the sort of energy balances that it yields under high heating load conditions. However, in the conditions of southern Europe the gross energy balance of north glazing in mid-winter is not substantially worse than that of south-facing glazing in some of the cool cloudy regions. For spaces which have low (or no) heating loads, but which require daylight, moderate sizes of north glazing can have cost and energy (and possibly, comfort) advantages over other orientations. - The relative disadvantages of west (and to a lesser extent, east) orientations are not clearly appreciated. These probably apply throughout Europe, though from the viewpoint of solar control the problem is more serious in the South, while in the North it is more an issue of unfavourable energy balance in winter resulting from relatively poor incidence of radiation and the high likelihood of overshadowing on glazing. Lack of choice on orientation, or an exceptional outdoor view, or other design considerations will require that openings of varying size are used on these orientations in some cases and it would be useful to study aperture characteristics for these situations. - The relative properties of tilted rooflights and atrium glazing of southern and northern orientation are an important topic for study, as these options emerge very frequently in the design of many spaces. It is suggested that for these configurations the issues of solar/daylight control should be considered in relation to possible night cooling 87

potential through longwave radiation to the sky, as well as in terms of the possible role of such rooflights in the ventilation strategy for a building. ι 3.2 Sunbeam Obstruction - Shading Scope Elimination of the direct beam and part of the diffuse component of the irradiance directed toward a surface, by obstruction along pre-selected parts of the daily and annual sunpath. Means -

Site topography and site layout; Trees and vegetation; Building form; Aperture components encompassing fixed and adjustable/retractable shading devices, internal blinds or shutters and combinations;

The means to be chosen need to be considered in conjunction with daylight and the desirability of solar heat gains at other times. Accepted

Know l edge

- The need for shading varies in time and with sky conditions, space functions, and design characteristics of the building enclosure. - The geometric aspects of shading are a function of geographic latitude, the azimuth and tilt of surfaces to be shaded, and time. - Generally, controlling solar radiation before it reaches a surface is a more effective way than the use of internal solar control devices. - F or spaces which may require solar access at times, adjustable or retractable means of shading are more effective and flexible than fixed (or permanent) devices, as they allow better control over diffuse as well as direct radiation and glare, and are likely to cause less overshadowing in the heating season; - Combinations between permanent and retractable components may offer advantages in many cases. Topics for Study - Southern Europe: - The high solar altitudes in summer in conjunction with prevailing clear sky conditions allow effective shielding from direct radiation with very modest overhang depths on southerly orientation. Combination of overhangs with awnings or adjustable louvres, or contraction of the exposed aperture using opaque panels, provides the flexibility to exclude sun at certain times while admitting solar gain in the cool period with the corresponding sunpath (e.g. exclude in September, admit in March.) However, for many spaces provision of adequate shading may be more important than ensuring completely unobstructed solar access in the heating season. Where all year solar control is required the choice between fixed and adjustable devices and the means for control become a function of daylight requirements and criteria. - The requirement for shading is not confined to the boundaries of a building, but will encompass surrounding outdoor spaces, F ig. 9: arcades, roof overhangs, balconies, verandas and pergolas have useful functions as shaded semi-outdoor spaces, and also as moderators of ground reflectance and of longwave radiation directed toward the parent building. While these functions may double-up with the shading of walls and openings on the parent building it would be wrong to compare these applications to single-purpose shading devices. Their environmental properties are more analogous to those of a conservatory in a colder climate. Trade-offs between design parameters can be sought depending on the desirability of such spaces and the relative importance between the requirement for shading and solar heat gain for a given building design. 88

FIG. 9 Shaded outdoor and semi-outdoor spaces are of crucial importance in S. Europe Topics for Study - High latitudes: - The relatively low solar altitudes and the need to avoid overshadowing in the heating season point strongly toward retractable external louvres for daylight and sun control, and simple internal blinds and curtains where solar heat gain is not a problem. - In cool and cloudy regions there is generally more advantage in preserving solar access to outdoor spaces than in providing them with shade. Shading from site obstructions and adjacent buildings During the heating season, overshadowing of windows and other glazed elements from site obstructions and adjacent buildings is a common source of considerable loss in useful solar gain. The effect becomes more pronounced on the lower floors of multi-storey buildings in dense urban regions, and with increasing geographic latitude. A recent study for France has expressed the effect of a wide range of such obstructions, including balconies and overhangs in the form of shading coefficients over the heating season for use in heating energy calculations [11]. In the south of Europe, high built density is a common feature of both traditional and contemporary urban development. In some cases the shading on building surfaces by adjacent buildings, in conjunction with shadows cast on walls and windows by balcony overhangs and vertical elements, may extend to most times of the year for the lower floors, Fig. 10. It may well be that close spacing could be desirable on an all-year basis in some cases, for example where the street direction is North-South. The value of the shading provided on the ground for pedestrians and circulation is another consideration; this urban canyon situation requires study of both the shortwave and longwave effects of reflected radiation, Fig. 11, [12]. There is scope for comparing the effect of different street widths to block height ratios with respect to both winter and summer requirements, including loss of daylight, and assessing differences in insolation and overshadowing between upper and lower floors. Trees and Vegetation as means of solar control The interest of using trees and vegetation as means of microclimatic control extends beyond their potential for shading and is an important topic for study. Depending on the design objectives, both deciduous and evergreen vegetation may be of interest, either with respect to the shading of outdoor spaces or for building facades. The latter is mainly of interest with respect to east or west facing surfaces. A number of computer programs are available to aid with the prediction of shadows cast by trees of various size and shape [13] [14]. These usually approximate the shape of trees as a combination of generic geometric forms and treat it as a solid. Another line of ongoing work is based on scanning and computer digitising of photographs of actual tree species for the purpose of characterising foliage and branch properties in the form of shading coefficients [15] [16]. It would be useful to have such typological characteristics. 89

F/U r

— *\

NE/SU canyui

\x\ ·

\ ­>

** „^ —

"



*

­

­

1 1 .SE/NW canyon N/S canyon

0

0,5

s t r e e t canyon r a t i o 2

i

4 5 Average of hour'; STREET SURFACE TEMPERATURE 800, 300, lüuj, M Û 0 , 1500, 1G00

70 ζ

ΝΛ canyon

60

..­.­:

ÍIW/SE canyoi

50 ■?''.

­

40 30

A

i '

:

H>1

NE/SU! canyon 1

20

-ί/U

canyon

10 0 0 0.5 1

I I 4 street canyon ratio

MEAH ŞHADIN.Ij PERCEHTA6E

FIG. 10 Shading of walls and ground by adjacent buildings

5

Latitude: 3 dey.Ν line ni TCcir : Septenber

FIG. 11 Street surface temperatures and shading as function of building height to street width ratio, [12].

Building Form As architects we will probably master the principles of bioclimatic design only when we acquire the ability to conceive form that is in itself an embodiment of explicit qualities of positive adaptation to, or exploitation of, ambient characteristics. In terms of solar control this may be through the way in which we might manipulate the roof, Fig. 12 in shape as well as properties; through the control of exposure on selected orientations or through an overall surface to volume ratio, Fig. 13; through plan form and sectional arrangement as means of protecting certain surfaces or spaces at certain periods; or by the extension of the plan in the form of balconies, courtyards and other semi­outdoor spaces that can play the dual role of shading device and usable space, Fig. 14. The questions of whether or when to glaze a courtyard or the gap in an L­shaped plan, thus creating permanent, or perhaps movable, glazed atria or conservatories, do not yet have' clear answers for southern European regions and deserve more study. 90

FIG. 13 Balcony extensions (top right, Le Corbusier)

FIG. 12 Roof shape (top left, Atelier Piano).

January t, July daily totals (lfctkUhAq«) and their ratio CO in Athens JfiH+ Ratio

FIG. 14 Effect of building shape on insolation (left, T. Stassinopoulos, MPhil research, AA School,London)

Aperture Components/ Shading Devices The range of possibilities is very wide both in terms of shapes, and with respect to materials, constructional techniques, and combinations between different types of devices and controls. Clearly, this is (and has been) an area both for demonstrating architectural ingenuity through context-responsive designs, Fig. 15, and for the production of standardised components for mass application, Fig. 16. It is all the more important to have fairly reliable comparative criteria and procedures for characterising and assessing performance of possible applications, and this should be an area of study (see also "shading coefficient" in next section). Increased recognition of the advantages of daylight as means of artificial lighting and cooling load reductions, as well as for its quality and psychological value suggest that it should be a major performance criterion in such assessments and specifications; this is touched upon by several recent studies [10][17]. 91

FIG. 16 Traditional multi-purpose louvred shutters

FIG. 15 Combination of fixed and movable shading at Asilo Sant'Elia (G.Terragni). Specific techniques, devices or issues which deserve further study (primarily for Southern European conditions) include the following : - Recommendations on appropriate typical geometric parameters of fixed shading devices for different parts of Europe as function of latitude and orientation and for different definitions of shading period and degrees of coverage of the shading requirement; effect of surface finishing of shading devices on daylight and longwave transfer. - Lightshelves, -fixed horizontal or tilted horizontal elements dividing the glazed aperture at selected points on the outside and/or inside, and which combine the function of shading with sunbeam reflection toward the ceiling for daylight penetration and distribution-, are of considerable interest as an alternative to, or combined with, overhangs on southern orientations. A recent study identified significant advantages for clear sky regions, Fig. 17, [18], [19]. - Slanted openings, a traditional technique which maximises skyview for a given glazed aperture while also providing shading, are a sort of natural form of lightshelf, Fig. 18; there is an ongoing European study testing a variant of this technique [20]. - Design characteristics of adjustable external louvres for solar heat/daylight control: materials, shapes, spacing to depth ratio; appropriate slat angles for solar admission, shading, light intensity control; types of controls and locations of sensors; possible improvements of optical properties by application of coatings, films, or mirrors on louvres [21]. - Evaluation of multi-purpose components combining function of solar control with movable insulation or other uses, for example, louvred shutters or blinds; "Beadwall" variants [22]. - Acceptable types of internal blinds or curtains that may be used for spaces of low solar control priority. 92

~ã a) g t u l n g ftuah wlih t h · lacada itcamnitndtd

·

·

·

E

Ρ?

b) glaring of claraalory window racaa« ad Irom lho lacada accvplaUa

·

·

7* Ughi anali daalgnad le allow pena •untight In winter through »law window (• d e p e n d on the thermal characteristics and geographic location of the building in which tlu\\ are installed. It was therefore necessary to develop detailed models of typical office buildings, of the cooling installations u n d e r s t u d y and their control s\ stems. In addition, comfort was characterized by means of an original criterion which has been the subject of a specific study. Finali?,, tests were conducted in jul.\ 1988 in the GAZ DE FRANCE e x p e r i m e n t a l building and a pilot operation followed up in s u m m e r 1989 in o r d e r to e x a m i n e the b e h a v i o u r of one of these systems in full scale operation.

135

3 - THE SYSTEMS STUDIED 3.1 - Representative buildings A typological survey was m a d e to select two reference buildings r e p r e s e n t a t i v e of a large s h a r e of m o d e r n office buildings. These two buildings differ only by their thermal inertia and the a m o u n t of solar input they receive. The first building has low inertia (100 kg/m2 according to the Th-B rules from the CSTB) and receives little solar i n p u t (the ratio of glazed areas to the floor surface is 4 %). The second has m o d e r a t e inertia (250 kg/m2) and receives m o d e r a t e a m o u n t s of solar i n p u t (10 %). All t h e o t h e r p a r a m e t e r s are identical. They include : 5000 m3 volume, 1500 m2 office area, North-South orientation, Coefficient of volume losses t h r o u g h walls Gl = 0,5 W/m3.°C, Controlled mechanical ventilation. A scenario of occupation and internal heat input was also established. 3.2 - Different cooling systems 3.2.1 - The reference In o r d e r to d e t e r m i n e the performances of different systems, a reference case was established : non-cooled building in which t h e o c c u p a n t s open the w i n d o w as soon as the inside t e m p e r a t u r e exceeds both 24°C and the outside t e m p e r a t u r e . The air change rate is in this case 7 volumes per hour. The windows remain closed d u r i n g the innocupied periods. 3.2.2. - Night-time ventilation One way to cool the building is to make use of its inertia and night-time cooling. Control is simple : as soon as the inside t e m p e r a t u r e exceeds the outside t e m p e r a t u r e , the m a x i m u m flow rate (4. 7, 12 or 20 vol/h were tested) is triggered. Otherwise, the flow rate is nul when the building is inoccupied and at 0,8 volumes per h o u r when occupied. 3.2.3 - Simple flux with direct humidification A first i m p r o v e m e n t of the system 2 is to cool the fresh air by direct humidification. However it is necessary to limit this humidification to avoid c o n d e n s a t i o n problems. The air humidification acts if : - t h e indoor air t e m p e r a t u r e exceed 24°C, - t h e indoor relative h u m i d i t y is less then 70 %. 3.2.4 - Double flux with humidification of exhaust When a system between supply the s u m m e r if control method,

of w a r m air heating with d o u b l e flux ventilation is used, a heat exchanger air and e x h a u s t air must be installed. This e q u i p m e n t can be used during a humidifier is fitted on the e x h a u s t air before the exchanger. A simple based on t e m p e r a t u r e thresholds, was developed. It is described in figure 1.

3.2.5 - Double flux with humidification of the exhaust air and fresh air In o r d e r to fully exploit the air network, it was deemed w o r t h w h i l e to test additional fresh air humidification. This a d d i t i o n a l stage is triggered if the inside t e m p e r a t u r e exceeds 24°C, providing t h a t relative h u m i d i t y in the officies is not greater than 70 %. 136

3.2.6 ­ Location s t a t i o n s Three types of c o n t r a s t i n g climate were selected according to hygrometry a n d t e m p e r a t u r e differences between day and night. The stations chosen are TRAPPES, AGEN a n d NICE.

non occupation Indoor temperature (Ti °C)

humidifier

flow air rate

heat exchanger

0

A ■19­

0 if Τ >T¡ 1 else ­20­ M

A ifTach>Te M else

0 if T s >Tj 1 else

M

Λ if Tat'h>Tt'

0 if T s >Tj 2 else

­21­ M else

occupation Λ ¿ ζ

M

1

A Λ if Taeli/Γι·

1

M else Ζ J

M

1 i f T >T¡ 2 else

Λ ¡Γ ΤίΚ·)ι>ΊΥ

Μ else

A off M on

Ts : Te : Ta e h :

suppU air t e m p o u t d o o r air t e m p air t e m p behind humidifier

0 nul 1 0,8 vol/h 2 = maximum

Figure 1 : cooling system control.

I

CRITERION FOR COMFORT ASSESSMENT

In an initial approach, based on the ASHRAE standard 55­81, a comfort zone was defined with the possibility of exceeding limit conditions during a certain number of hours. This /.one is defined b\ an effective temperature limit (operative temperature at 50 % relative humiditv ) of 2(>°C. a relative humidity limit of 70 % and an absolute humidity limit of 13 g/kg of dr> air. A simple calculation of the n u m b e r of hours during which t h e threshold of 26°C is exceeded gave an initial evaluation of systems. This m e t h o d takes a c c o u n t of the d u r a t i o n of discomfort but not its degree.

137

For this reason, a criterion taking account of both p a r a m e t e r s ( d u r a t i o n and degree of discomfort) based on the work conducted at the JB Pierce F o u n d a t i o n (P. GAGGE) and the Technical University of Denmark (P.O. FANGER), was chosen. At any m o m e n t , the effective t e m p e r a t u r e , the PMV (Predicted Mean Vote) and t h e PPD (Predicted Percentage of Dissatisfied) are d e t e r m i n e d on the basis of operative t e m p e r a t u r e and humidity. A mean discomfort d u r a t i o n is calculated over the entire hot season by integrating the difference (PPD ­ 0.10) when PPD is greater t h a n 0,10 (the comfort zone corresponding to a PPD value less t h a n 0.10). B y separing warm and cold mean discomfort d u r a t i o n (WMDD or CMDD), it is then possible to make a more a c c u r a t e classification of the systems. 5 ­ SIMULATION RESULTS The tool used for this study is the ASTEC 3 software package, an algebrodifferential system solver developped initially for the description and simulation of electric circuits. The models are t h u s described in the form of electric circuits using t h e electric t h e r m a l analogy. The different modules developed for this study are used to e x a m i n e in detail the dynamic b e h a v i o u r of the systems in question. The simulation results have been studied using the mean discomfort d u r a t i o n (MDD) criterion. Energy and w a t e r needs have also been evaluated. As an exemple, figure 2 indicates the WMDD value for Trappes site according to the building and system type and the o r i e n t a t i o n (South or North facing).

1 η

— o

i-t

u

^ c u

ViΊι '

il

ι>­­4 •

23

s

Ύ 29

ι

3b

\ [

Ι 1/ ; ■ ­

30 4J

'

Ι '· ι

■:

:J

Ι' Ι ?!

!¿

Ι 28

Ì τ Ι

4

I

I j

6 ;'j

¡ί m

Ϊ

~~τ

! 24

Π

*·>­_

7

-- i 2

­l·­ — i

1

't

Γ

i

ι

1 ow.

S

1

f—

Ululi .

Ν

­

Γ

.ol/t,

Ν

i

­

­

Ι

! _

—r _ " I '

ι

y

Ι

~>

14

Ι

ί»

l'i ίο

­

Ι

ι

­

|'



Ι ο

ι ;

ι

4

1

In

\

i

Ι

''

ί

f igure 2

Warm mean discomfort duration (hours) for different cooling systems, airflow rate and buildings at Trappes

For t h e t h r e e w e a t h e r stations, the system 5 which is the most c o m p l e t e is also the most efficient. Systems 3 and 4 give close results. The system 2 doesn't give sensibly better results t h a n the reference case and will not be further considered. Detail results are the following : Trappes : For the low inertia building and an efficient solar protection, systems 3, 4 and 5 need 7 vol/h air flow. For a m e d i u m inertia building systems 3, 4 and 5 can be used with a 4 vol/h air flow if the solar protection is good. Agen : The possibilities are the same as above if the solar protection is very efficient, and if internal heat gain are low. If not, low inertia building need a 12 vol/h air flow and medium inertia building a 7 vol/h air flow. Nice :

A 7 vol/h air flow can be used only for the m e d i u m inertia building with very efficient solar protection and low internal heat gain. For o t h e r s buildings, a 12 vol/h or 20 vol/h air flow is needed.

138

6 ­ TESTS IN THE GAZ DE FRANCE EXPERIMENTAL B UILDING The building comprises 25 ¡noccupied a p a r t m e n t s on five floors, in which thermal e q u i p m e n t and systems are tested. The main aim of the tests was to verify the simulation results for the indirect e v a p o r a t i v e cooling system. On three floors of t h e building, the air network was used to d i s t r i b u t e air from a central unit equiped with a d o u b l e flux system with e x h a u s t air humidification at a rate of 4 office volume/h. The t o p a n d b o t t o m floors were not cooled. The same occupation scenario as for t h e numerical s i m u l a t i o n s was a d o p t e d for all floors. A control similar to t h a t of the s i m u l a t i o n s was applied. These tests gave first informations on the comfort improvement, and on t h e r u n n i n g characteristics of the air cooling system (humidifier and air exchanger efficiency). 7 ­ PILOT OPERATION A pilot operation carried out at Senlis near F^aris has been followed u p d u r i n g the s u m m e r of floor area of 3000 m fr ot \\19KÍ). >K). The I tie three level building has a Hoor m".. The The floors are free of inside walls, except for some rooms situated in the corners of the building. Two indirect evaporative systems have been settled up respectively for the north and the south part of the building. Air t e m p e r a t u r e and humidity were mesured in the air cooling units and in some points inside the building. Though the o u t d o o r air t e m p e r a t u r e s have been sensibly w a r m e r than for a mean summer, indoor air t e m p e r a t u r e remains generally between 20°C and 26°C, with a daily variation less than 26.4 C LiaftON­IIOMC M t

­ « ­ DOUBLE βΤΑΟΕ

­ · ­ « H O L E BTA OE DIRECT

Figure 7 :Percentage of possible working hours

platc.the tubular and the rotary type coolers.Detailed description is given in [24]. The threshold value for the operation of the system is that the indoor air wet bulb temperature should be lower than the outdoor dry bulb.In practice.thc indoor wet bulb should be lower than 21 C According to the reported data ,[26],the performance of the system is quite satisfactory. Energy savings of up to 60 percent compared to compression refrigeration systems may be achieved in hot dry regions [24].However,the efficiency of the system is strongly influenced by the wet bulb temperature of the outside air. As indirect evaporative systems do not add moisture to the building,no humidity control is required for their opcration.Corrosive components should be avoided for maintenance reasons.Effective filtering is necessary to reduce the number of dust particles. Where the ambient air is too high.a two stage evaporative system can be used.This consists of an indirect cooler coupled with a direct one ,or/and an indirect cooler.They may be coupled with a refrigerative A/C unit. There are many applications of these systcms.especially in California,and there is a number of established manufacturers.also in USA.Energy savings for such a system have been reported at close to 50 percent compared to an equivalent A/C system [24]. The association of a direct with an indirect evaporative cooler results in lower temperature threshold values reducing thus the possible working time of the equipment.However a double stage cooler offers lower dry bulb temperatures than a single stage,resulting thus in increased indoor comfort levels.Figure 7 shows the percentage of possible working hours of a single ( + ),as well as of a double (*),stage cooler as a function of the period where cooling load exist^A performance statistics has been derived after extensive calculation for 14 South European locations.Similar data are also presented in a monthly basis in figure 8 for Athens.As it is shown single stage direct evaporative coolers can operate 5 to 20 per cent longer than the double stage equipment. However the use of direct evaporative coolers increase the indoor humidity levels,and decreases comfort.In figure 9 the range of the relative humidity of the exit air from a single direct and a double stage cooler in Athens is givcn_As it is shown the use of double stage coolers should result in lower indoor humidity levels.Therefore the percent of time that direct coolers will deliver air cool enough to maintain the effective space temperature as specified by the user .comfort indcx,is limited.Figure 10 shows the mean monthly comfort index achieved in Athens using a single direct and a double stage evaporative equipment.lt is clear that under these climatic

147

RELATIVE HUMIDITY RA NGE

ATHENS ­ COOLING NEEDS ­ WORKING HOURS FOR SINGLE A NO DOUBLE COOLERS

May

JufM

July

A uguat

BINÓLE A ND DOUBLE STA GE COOLERS

•0­70

Saplaaibar

RKRANGE

MONTH aafl D O U B L E a m a «

Ì Z 3 a m o L i STA SE

I SINGLE STA GE

BS3 O O O L I N Q N E E O C

Figure 8 : Needs and working hours of the coolers

E 5 2 DOUBLE STA GE

Figure 9:R.H.ourput from single and double stage coolers

conditions direct evaporative coolers are of limited ability to provide effective air conditioning.The mean summer comfort index for 14 South European cities resulting after the use of direct evaporative coolers is given also in fig. 11 as a function of the 1 % wet bulb temperaturcAs it is shown higher w.b. temperatures correspond

ATHENS­COMFORT INDEX

COMFORT INDEX

SINOLE A ND DOUDLE STA GE COOLERS COMFORT INDEX («1

00

COMFORT INDEX ( « )

* Nh,

\ * %

M

%

ν

*

M

s

«s

's*

70

7-

»

ao

*

*

s

\.

JULY

*^\

MONTH 20 ­ SINOLE STA GE

­ DOUBLE STA GE

Figure0: Mean monthly comfort index in Athens

20.« 21

21«

22 22.» 23 23.6 24

24.3 23

26.3

M

1 % W.B. TEMPERATURE

Figurell. Mean monthly comfort index for 14 locations

to lower comfort indexes Future Research Actions. A future research action on evaporative cooling should be focused on six main areas :

1.Collection and classification of relevant climatological data for Europe. 2 Assessment of the utilisability and benefits of evaporative cooling techniques for European conditions. 3.Development and validation of software for the calculation of the performance of selected systems and techniques. 4.Experimental study of promising passive techniques and assessment of the main design parameters. 5.Development of appropriate coolers in conjunction with European industry. ó.Production of a handbook describing the applicability of evaporative cooling techniques and systems in Europe. EARTH COOLING During the summer,the soil temperature at certain depth is considerably lower than the ambient

148

temperature.Therefore ground offers an important source for the dissipation of the buildings excess heat.Seasonal variation in earth temperatures decreases with depth,moisture content,soil conductivity and with the surface covering.The relative ranges of the soil temperature as function of depth for various types RANGE OF THE SOIL TEMPERATURE of soil is given in figure 12 [27]. At the same time ground can contribute to the *&I reduction of the heat gain from the exterior,offering 10 effective solar and heat protection. fc. There are two main strategies for the dissipation of I» ·■■:; the heat to the ground.The direct earth contact 10 cooling which involves partial or total placing of the t • · · buildings envelope in direct contact with the soil,and . * · ·· the earth pipes technique which involves the use of a buried pipe where air from the building or from DEPTH (U) the outside is circulated through the pipe where it is Figure 12 : Variation of the soil temperature with depth precooled and then is brought into the building. Direct Contact B uildings The transfer of heat from the building to the earth through the walk ,the floor and possibly the ceiling is a well known technique. Earth coupled techniques have been used at different times in history and in different parts of the globe.Important underground dwellings.villages and communities have been also developed in the Mediterranean region,[28­31]. During the last years the use of the earth as a heat sink for climate control has become popular.Estimates of the number of earth sheltered houses in USA in 1982 range from 4000 to 8000 [32].In Europe there is an large number of one or two story buildings which are set into hillsides,placed partially or completely below grade. Knowledge on the performance of earth sheltered buildings comes mainly from measurements of traditional buildings,[33],from monitoring of new constructions,[34­35],as well as from theoretical analysis,[32].Energy gains of about 50 to 90 percent are reported for various monitored earth sheltered buildings,[34­35]. Earth contact buildings offer various advantages i.e limited infiltration and heat losses,solar and heat protcction,reduction of noise and vibration,fire and storm protection and improved security Also they present important environmental and land use benefits while their maintenance and operation cost is low. However they are not free of disadvantages.Inside condensation,slow response to changing conditions,poor daylighting and poor indoor air quality are frequent problems.In addition high cost reduces the potential for a wide scale constructions of such buildings.

Buried Pipes Underground cooling tunnels is a concept that can be traced back several centuries.Applications of these techniques at different times and in different parts of the world are described in [1,36,37]. The concept involves the use of a metallic or a PVC pipe buried at 1 to 3 m depth.Ambient or indoor air is delivered inside the tubes where it is precooled and then is delivered to the building.When outdoor air is circulated into the pipes the system is characterized as an open loop system while when indoor air is recirculated from the building through the tubes the system is known as a close loop system. The performance of the buried pipes is a function of the inlet air dry bulb temperature.the ground tcmpcraturc.thc thermal characteristics of the pipes and soil as well as of the air velocity,the pipe dimension and the pipes depth. Techniques to estimate the efficiency of such systems have been developed by various authors and a review of the existing methods is given in [38]. Knowledge on the topic is coming mainly from single pipe experiments [39­46],as well as from theoretical analysis.Results reported in [47],show that for a 30 m long PVC pipe buried under shaded soil at 1.5 m depth,the maximum temperature drop inside the tube was 18 C,(Fig. 13).However the temperature reduction is time variablc.The inlet and exit air temperature for 120 hours of continuous operation,taken in the previous

149

BURED PIPES PERFORMA NCE • U T A MI OUTLET

BURIED PIPES PERFORMA NCE

T B R M n M TVMrwuTum

IMJIWIII — 40

M «0

^

WLET

\__

m

ae so

OUTLET

m

Ν

MM/

it ■OmiMMMT

*

TME DAYS

Figure 13:Inlet and outlet temperature from a buried pipe

MLfT

­*­OUTVTT

Figure 14:Continuous operation.Inlet and outlet

experiment are given in fig 14,[47].The exit air temperature was significantly lower than the inlet one,during daytime and significantly higher during night.is indicates that buried pipes can be used for precooling as well as for preheating of the building. Special problems related to this technique is the limited potential for dehumidification,reducing thus the possibility for latent cooling.Condensation inside the tubes can occur only with very low air flow and high ambient dewpoint temperatures.Also in damaged tubes water,is possible.to enter into the tubes.Moisture accumulation can lead to biological growth and resulting odour problems.However no such problems have been reported.

Future Research Actions Suggested future research actions on the field of ground cooling are the following: l.Collection of relevant climatological data,especially on ground temperature for different types of soils. 2Λ. review of the literature on earth sheltered buildings which could be followed by selected short term field studies and occupant interviews. 3.Development of an evaluation method for the prediction of the performance of partially covered buildings and validation of the method. 4.Experimental activities on the technique of buried pipes under different type of soils,materials etc. 5.Development of a model for the prediction of the efficiency of buried pipes and validation of the model.Studies on the control of the systems and on the coupling of the system with the building and with conventional cooling systems. ó.Practical guides regarding the application of those systems and dissemination of information.

RADIATIVE COOLING Radiative cooling is based on the effect when heat is lost by a body due to its long wavelength radiation to the night sky .The net radiant heat loss from the body is the balance between the emitted energy flux and the absorbed incoming atmospheric radiation.Atmospheric radiation depends mainly on the level of cloudiness.Under cloudy sky conditions the atmospheric radiation reaches its maximum minimizing thus the net radiant heat loss.Vapour content and increased aerosol concentration tend to increase also the sky radiation [48].However the net energy loss from a body is the balance between the net radiative loss and the convective heat exchange between the radiator and the ambient air. Convective heat transfer from the radiator is a function of the wind speed near the radiator and is proportional to the temperature difference between the radiator and the ambient air. Radiative cooling techniques were well known from the ancient years and they were used to produce ice and to cool the buildings. During the last years radiative cooling has been investigated for a variety of building cooling methods. Existing cooling techniques can be classified as passive or hybrid. Passive systems, mainly involve the use of the building roof as the radiative component, while hybrid systems involve the use of special metallic surfaces characterized by high emissivity in the longwave range. Important techniques for passive radiative cooling is the "white painting"of the roof the use of the concrete roof associated with operable insulation and the "Skytherm" system.

150

Passive Systems The colour of the roof influences the thermal performance of the building significantly because it governs the absorption and reflection of the incident short wave radiation during daytime and the emission of longwave radiation during night time. Painting the roof white increases the reflectivity of the roof to the solar radiation reducind thus its temparature. This technique is traditionally followed in the Mediterranean region and more especially in the Greek islands. Measurements of the performance of this technique reported in [49] give a cooling potential of 0.014 KWH per square meter per day. Experiments also reported in [50], using different roof colours, give a temperature difference of 3 C and 1C ,when measured 0.1m below the roof and 1.2 m above the roof, respectively for grey and white roofs. However, the effect of colour is more important with light structures than with structures of high thermal capacity. The main problems associated with this technique is the high rate of heat loss during the winter imposing for a new change of colour, the high indoor temperatures during daytime in the summer and that is effective only in single storey buildings. In conclusion, this technique is of poor performance and it should be considered as a solution only in warm regions. Exposing the cold storey mass to the sky during the night while protecting it during the day by a movable insulation optimises the potential of radiant cooling. Operable insulation can be in the form of horizontal movable panels or hinged panels positioned vertically during the night. Measurements of the performance of this technique reported in [49,50] give a cooling potential of 0.266 KWH per square meter per day. The minimum recorded temperature of the surface was 14 C while the maximum was 19 C . Also the minimum indoor temperature was 15.2 C and the corresponding maximum 22.4 C [49]. The main problem associated with this technique is that is effected only in single storey buildings while the horizontal operable insulation requires storage spaces for the night,and the hinged panels cover an area of the radiant roof corresponding to their thickness Also it should be pointed out that the system require important man or mechanical operation. The "Skytherm" system proposed by H.Hay involves the use of water bags which are placed on the roof Λ movable insulation is placed above the bags during day to keep the solar heat away from the water and is removed during night where the water loose heat by convection,radiation and evaporation and thus the water cools the living spaces. Various other types of architectural integration of this system are proposed in [51]. Knowledge on this topic comes from experiments of Hay,[52],at Atascadero House in California.The results of this experiment are reported in [53].It was observed that on a typical summer day.the variation in outdoor temperature was from 13 C to 34 C,while the corresponding indoor temperature ranged from 21 C to 23 C only. However the Skytherm system present important limitations.Some of these are : 11 is applicable only to one storey buildings or to the upper floor of multistorey buildings A deterioration of the plastic with time,is observed reducing thus its transparency. There is an accumulation of dust and The cost of the system is important.

Hybrid Systems Hybrid radiative cooling systems involve the use of specialized metallic longwave radiators which can reach temperatures below those achieved when the roof is cooled directly.The radiator should be lighweigh and possibly insulated underneath to avoid heat flow to the radiator from the roof below.Th η e operation of the system involves the flow of a transfer medium,water or air.flowing above .under or within the radiator which is cooled and then is used directly or it is stored in order to cool the building. Storage can be achieved circulating the cool water into the concrete floor.which serves as combined cold storage and radiant convective cooling pancl,[54].Cool water can be also stored in water tanks,[55],for later use. The potential of metallic radiators for radiant cooling is a function of the possible depression of the radiative surface tcmpcratureTh.e maximum depression is achieved under stagnation conditions,!.e when the radiant heat loss is not used.The corresponding temperature of the surface is refered as stagnation temperature and

151

express the potential for radiant cooling for any climatic conditions. MEAN AND MEAN MAXIMUM The mean temperature depression under stagnation conditions for 18 Mediterannean locations is given in LISBON PORTO figure L5.Stagnation temperatures have been VALENCIA BARCELONA calculated using the method proposed in [56].As it is MADRID PALMA shown under mean climatic conditions the expected PERPIGNAN MARCE ILLE temperature depression .under stagnation NICE conditions,rarely exceeds 3 C.In the same figure the TORINO MILANO possible maximum temperature depression is also ROME NAPLES given.In this case the corresponding stagnation PALERMO SALONIKI temperatures have been calculated using optimum ATHENS TRIKALA climatológica] conditions,!.e clear sky ,low wind IRAK LIO speed etc.The calculated maximum depression temperature range from 5 to 7 C. Figure IS : Mean and maximum stagnation temperatures Improvement of the performance of metallic radiators can be achieved using high emissitys selective surfaces and infrarent transparent windscreens to reduce convection from the surface. Selective radiators present a high emissivity in the 8-13 microns wave band while are highly reflective above and below this wavelenght band.However selective radiators are advantageous only when its temperature is lower than the effective sky temperature.In this case the intensity of the atmospheric radiation above and below 8 - 1 3 microns is higher than the radiation emitted by an ordinary surface in these spectral region.However when the radiator's temperature is higher than that of the sky the emitted from the surface radiation exceeds that of the atmosphere at all wavelenghts,and thus selectivity does not offer any advantage regarding ordinary radiators. Experiments with selective surfaces,and more especially with anodized aluminum of 4 and 10 microns,aluminum with sodium silicate layer and aluminizcd tediar sheet,described in [57-61] report temperature's dempression from 5 to 17 C.Comparisons with ordinary surfaces have shown however that the difference between them is negligible Also when the temperature depression is important,condensation occurs on the surface and thus the surface lose its selective radiative properties. Wndscreens transparent to infrared radiation decrease the convective heat from the ambient resulting thus in lower radiator's temperature.The material commonly used is polyethylene without U.V. inhibitors.Polyethylene is characterized by a transmissivity to longwave radiation equal to 0.75 ,a reflectivity equal to 0.1 and an emissivity equal to 0.15 [62].Polyester and fiberglas films have been tested also without sucess [57]. Experiments ,[62-63],report that a radiator covered by a windscreen was 3-4 C lower than the corresponding temperature of an ordinary radiator .However when the temperature of the polyethylene film drops below the ambient dew point ,dew is formated over the windscreen ,and thus the transmissivity of the polyethylene to the infrared radiation is reduced significantly. STAGNATION TEMPERATURES

Future Research Actions The suggested future research in the field of radiative cooling may be classified in the following actions : Collection of the sky temperature for various sites in rope in order to realize a sky temperature map.Improvment of empirical models to calculate the sky temperature. Development of an evaluation method for the performance of radiators taking into account condensation,convection,dust covering and thermal coupling with the buildingExperimental validation. Development of low cost .efficient selective surfaces Integration of radiative systems in the building design Guideline on the use and the sizing of these systems. CONCLUSION The state of the art on the natural cooling techniques is presented.The more important techniques and componentsof evaporative.earth contact and radiative cooling are presented and their performance is discussed.Future research actions are proposed for each topic.

152

Acknowledgments The author wishes to thank Prof. P.Koronakis and Dr A.rgiriou for his valuable contributions. REFERENCES l.Fanciotti and G.Scudo : Proc. Int. Passive Hybrid Cooling Confere nee. Miami Beach.p.l79ABowen,E. Clark and K-Labs (Edts),1981. 2.Predicasti Inc : Predicasti Forecasts.Issue 108,4th Quarter ,Refrigeration and Air Conditioning Equipment Data,1987. 3.United Nations : International Trade Statistics Yearbook,Vol II.Trade by Commodity.1980-1985. 4CJ.Andrews : The Energy Journal.10,3,107,1989. 5.D.A.Montgomery : Passive Solar Joumal,4,1,79,1987. 6.A.Moffat and M.Schiller : Landscape Design that saves Energy.New York. William Morrow and Company,1981. 7ABowen : Passive Cooling HandbookAS/ISES.1980. 8E.Duckworth and J.Sandberg : Bulletin of American Meteorological Society ,35,198,1954. 9H.AkbariJ.Huang,P.Marticn,L.RainerA.Rosenfeld and H.Taha : ACEEE Summer Study on Energy Efficiency in Buildings.1988. 10.W.A.Cunningham and T.L.Thompson : Proc. PLEA 86,Pecs,p.S-23,1986. ll.G.Mignon.W A.Cunningham and T.L.Thompson : U.S. Department of Energy Contract No.DE-FG02-84CH10205,ERL ,1985. 12.B.Givoni : Proc. PLEA 88,p.521,Porto.E.de Oliveira Fernandes and S.Yannas (Eds). 13.D.W.Abrams : Low Energy Cooling.Van Nostrand Reinhold Co. Inc.1986. 14 J.Yellot : Advances in Solar Energy,p.241,1983. 15.S.P Jain :Building Digest No 124,C.B.R-I.,Roorkee India.1977. ló.A.B.Thappen : Refrigerating Engineering ,163,1943. 17.L.H.Holder : Automatic roof cooling.Ail Showers Company,Washinghton,DC,2,1957. 18.S.M.BIount : Ind.Exp.Prog.Facts for Industry Ser.,9,1958. 19.HR.Hay and IJ.Yellot : ASHRAETrans.75,1969. 20S.PJain and K.R.Rao : Building Science ,9,9,1974. 21 AL.Pittinger,W.R.White and K.I.Yellot : Proc. Sec. Nat. Passive Solar Conference AS/ISES,p.773,Philadelphia,1978. 22.M.S.SodhaA-K.Khatry and M.A.S. Malik : Solar Energy,20,189,1978. 23.M.S.SodhaAKumarASingh and G.N.Tiwari : Building Environment 15,133,1980. 24 J.Watt : Passive Solar Journal 4,3,293,1987. 25.ASHRAE :Handbook of Fundamentals. 26.D.Pescod and R.K.Prudhoe : Telecommunications Journal Australia,30,2,1980. 27.B.Givoni and L.Katz : Energy and Buildings,8,15,1985. 28F.Hazer Proc. Conf. T h e use of earth covered buildings" F.Moreland (ed),US GPO 038-000-00286^,1975, pp 21-36. 29R.Cole and R.Kennedy. 5th NPSCJ.Hyes and R.Snyder,(eds),S/ISES,Newark,Dclawarc,pp.704-706,1980. 30.S.Baggs : Underground Utilization ,a Reference manual of selected works.T.Stauffer,(ed),pp.573-599,1978. 31.0.Newman,ed CIAM 59 in Otterlo,Alec Tiranti,Ld,1961. 32J.C.Carmody,G.D.Meixel,K.Labs and L-S.Shen : Advances in Solar Energy,2,297,1985. 33.EChronaki :Ph.D. Thesis,University of Thessaloniki, 1983. 34.D.Carter : Underground Space 1,317-323, 35.WJ.Rrvers,B.Helm,W.D.Wardc and W.Grondzik : Ibid 1,ρ.126,1981. 36.B.S.Saini : Building Environment : An illustrated analysis of problems in hot dry lands.Angus and Robertson,Sydney,1973. 37M.Bahadori : Scientific American,238,2,pp.l44-154,1978. 38ATombazisA-Argiriou and M.Santamouris : Int. J. Solar Energy.In Press,1990. 39M.Santamouris : DEA Report,I.N.P.Grenoble,1981. 40CEFrancis : Ibid 1,1981. 41J Claeson and A.Dunand :"Heat extraction from the ground by horizontal pipes'.Swedish Council for Bilding Research,Dl,1983. 42.A.l_T. Seroa da Motta and A.N.Young : Proc. INTERSOL 85,E.Bilgen and K.G.T.HoUands (eds),p.759,1985. 43.G.Schiller:M.Sc. Report,L .B.L .,1982. 44.Abrams,.Donald and C.C. Benton : Proc. of the 5th National Passive Solar Conference AS/ISES,Amherst,1980. 45A.S.Dhaliwal and D.Y.Goswavi : Proc. of the 6th Annual ASME Solar Energy ConferenceAS.M.E.,1984. 46.Nordham and B.Douglas : Proc of the 4th National Passive Solar Conference.Kansas City,1979. 47.M.Santamouris : Report on the 205/85 Energy Demonstration Project,D.G.17,EEC,1989. 48.G.Clark : Ibd l,p.682,1981. 49.B.Givoni :Ibid l,p.279,1981. 50 B.Givoni and M.Hoffman : Build. Intemational,6^25. 51.K.L.Haggard : Solar Energy,19,403,1977. 52.H.Hay : Proc. Conference Passive Solar Heating and Cooling,Abuquerue.New Mexico,1976. 53.K.L.Haggard : Research evaluation of a system of natural air conditioning.port by California Institute ,1975. 54BJuchau : Ibid l,p.256,1981. 55W.Land : Ibid l,p.274,1981. 56.S.IO and N.Miura : J.of Solar Energy Engineering,l 11,3,251,1989. 57J.T.Pytlinski,G.R.Conrad and H^Connell : Alternative Energy Sources,P.101,N.Veziroglou ,(ed),1983. 58F.Sakkal : Ibid 29,p.483,1979. 59W.CMllerJ.O.Badley : Ibid 29,p.480,1979. 60.S.Catalonotti et al : Solar Energy,17,83,1975. 61B.Undro and P.G.Mc Cormick : Int. Heat Mass Transfer.23,613,1980.

153

Use of an Expert System for Passive Cooling Design of Commercial Buildings G. Pecorella, A. Santamaría, G. Silvestrínl CNR-IEREN - Viale delle Scienze - 90128 Palermo - Italy

Abstract The work in progress of an Expert system called ISOLDE that is being prepared in a specific Task of the International Energy Agency is presented. This program will give intelligent user support on energy use and thermal comfort during the design process of commercial buildings through general advices, simplified methods, detailed simulations. In particular the paper focuses on the passive cooling approach covered by this tool.

1. The Isolde program: an expert system for energy advice on commercial buildings In Task XI (Passive and Hybrid Solar Commercial Buildings) of the International Energy Agency a group of researchers is working on an expert system called ISOLDE (Integrated Knowledge Based Solar Design Tool) that shall primarily be a tool for interactive acess to the results and experience obtained from the work in the Task. The research is still at the beginning and the first prototype will be available in October 1990. The architecture of the system, shown in fig.l, is divided in three parallel parts: General advice, Case oriented advice and Analysis. The General advice part can be seen as the educational part of the system and includes an overview of possibile options to be used at the first steps of a project. Rules of thumb on different systems will be provided at this stage and the possible side-effects between cooling, heating and daylighting will be emphasized. To increase the usefulness of this part

ISOLDE Manager

General Advice

Case Oriented Analysis

Calculation / simulations Results evaluation

Video interface

Documentation

Fig. 1 General architecture of ISOLDE 155

it will be possible to access to a video system in order to show slides and illustrations mainly taken from the experience in monitoring and simulations gained in the IEA Task. The Case oriented analysis will give the performance of a specific project proposed by a designer, given the climatic data and the building description. The results from simplified methods will be presented as graphs showing the sensitivity of choosen building parameters. The Energy analysis will be performed, if necessary, by activating a simulation model; the results will then be evaluated giving more detailed informations to the designer. ISOLDE is implemented on PC using the shell CRYSTAL. Simplified methods and simulation methods are standard tools that will be activated from a batch file when CRYSTAL closes down. When the calculations are finished the batch file will restart CRYSTAL In order to lower the computer answer time and to allow the separate work of the different research groups involved in the ISOLDE project it was decided to create many small knowledge bases (fig. 2). The structure includes a first strategy evaluation knowledge base, three strategy knowledge bases (heating, cooling and daylighting) and a number of principles and system knowledge bases. Based on the climatic region of the site analyzed and the building size (floor area, number of stories ) and use (office, hotel, hospital, school, shopping-center, storehouse, sport center), Isolde will provide a simple ranking of strategies. In this paper we will describe how the cooling strategy is organized.

2. Cooling strategies If cooling is a recommended option in the case analyzed, more specific climatic data are required. According to the average summer weather data an indication of possible cooling strategies is provided (Tab. 1). Three informations (not an issue, usable, recommended) will be provided by ISOLDE according to the input data.

Summer Data

Shading

Natural Forced Ventilation Ventilation

Evaporat.

Daily AT (C) T. max (C) Rel. Humidity (%) Wind Velocity (m/s) Solar Rad. (Wh/m2 d) Tab. 1 Cooling strategies proposed in relation to summer climatic data 156

Underg.

Roof Pond

Common Rules

Knowledge bases

Screen pointers

pointers ISOLDE Manager Strategy evaluation

ISOLDE Manager Heating

ISOLDE Manager Cooling

ISOLDE Manager Daylighting

ISOLDE Manager Ventilation

ISOLDE Manager Atria

ISOLDE Manager

ISOLDE Manager Window Air Collector

I Video Interface

Fig.2

ISOLDE architecture (General advise part).

For each principle or system selected the knowledge base activated will provide: • System/principle definition 157

• • • •

Advantages/disadvantages Side effects to other strategies/principles/systems Design Advice Rules of thumb for design

Climatic data combined with the building type permit to give some general recommendation. In Fig. 3 for five different climates the cooling strategies are explained for the case of office buildings (De Herde, 1990). Buildings at this stage are considered in relation to the level of internal heat gains and of permeability (a permeable building allows a large amount of heat to flow inside, or outside, the envelope). A bold square has been drawn each time a principle or a strategy proves to be relevant according to climate and daytime. In this way all possible cooling strategies relevant to the case analyzed are presented. At this point it is possible to have more detailed informations on specific strategies concerning general advices,"advantages", "disadvantages", possible "side-effects" on other cooling strategies, on heating or on daylighting, rules of thumb for design. In the following paragraphs some of the cooling strategies analyzed in Isolde are described in more detail. 2.1. Ventilation For the different cooling startegies the first step has been an analysis of the literature in order to identify rules of thumb, correlations and simplified methods available. In the following paragraph a review of tools identified to quantify natural ventilation in buildings is presented. A more detailed analysis The calculation of airflows through buildings is difficult and cannot be done with precision. Uncertainities on site wind speed, on surrounding landscape perturbance, on pressure coefficients are great. A growing amount of wind tunnel experiments however has recently permitted to define better correlations. A lot of research is at present focused on the improvement of the definitions of the parameters involved in natural ventilation. Simplified methods and simulation models to be used for the evaluation of buildings ventilation are already available. In many simplified calculation methods it is possible to evaluate the effect of different air changes. This is the case of the procedure to calculate summertime temperatures in buildings proposed by Loudon (1968). Baer (1983) presents a simplified method that allows to calculate the internal temperature with a fixed ventilation rate. A manual calculation method that permits the evaluation of night ventilation is presented by Baker (1986). A semi-empirical thermal analysis method has been developed by Mathews (1986, 1989). The same approach has been followed in a program, called QuickTemp, that has been validated against 39 separate commercial and residential buildings (Joubert, 1989). Mc Farland (1989) calculates ventilation cooling using curves function of climatic data and building characteristics. A simplified model for the prediction of air flow im multistorey buildings is proposed inFeustel(1989). A model that has been used to illustrate the effects of wind on air change rate in relation also to terrain roughness is presented in Liddament (1988). Etheridge (1988) has developed non-dimensional graphs that can be used to predict ventilation rates in buildings. 158

OFFICE

Oslo

Brussels

Zurich

Rome

Dallas

D/N

D/N

D/N

D/N

D/N

BB BB BB BB

BS fflB

■B BB

BB ■B

US

KB

BB ■B fflB BB

BB BB BB BB

BB BB EBB BB •

BB BB

■B SB

B B

BB

BB BB BB

EBB

BB

EBB

BBS

BB BB BB

US BB

SB HB BB

BB

BB BB BB SS SB SB

BB BB BB BB BB BB

HB ■Β BB SB

■B fflB ■B

■Β

BB ■B

CB CB

CB CB

GS CB

BS CS

EH BU SB HB

EH BB BS

BH BH ■Β

BH BH BB HB

BB SB

SB SB

ss

BB

BB BB SB

sa [UB υ is m io«

BH SH BH

SOLAR CONTROL Window protection Movable devices Permanent devices Vegetation Special glaring Building protection Orientation Earth protection Vegetation ♦ obstacles Roof addition Environment protection Vegetation effect Shading systems Compactness H E A T A VOIDA NCE Reduct, of transm. gains Surface treatment Insulation Thermal inertia Shape and compactness Reduct. of infiltrations Zoning I N T E R N A L GA INS RED. Reduct. of lighting gains Reduct. of casual gains NAT. V E N T I L A T I O N Temper, gradient effect. Wind pressure effect

BB BB BB BB BB BB BB ■1] am ÜJLB

Outside principles N A T U R A L COOLING rhysiolog. cooling

fflffl

Air cooling

SB

Evaporative cooling

BB

Air ducts in earth Envelope cooling

ss ss

Night, radiative cooling Convettive cooling Conductive cooling ■ "] hiph perm cab UJ high uit.gains

Fig. 3

[■ LO

HB

Œ3B

ESB

SB SB LULU SB high perm low int. giΣΕ «

EEB

kJW

low permeab. high inL gains

EBB

■B BB SB BB

BBS

ms

rs

ms

mm

1



mm

pernieab. low inL gains

Cooling suggestions for office buildings in different climate

Owens (1987) presents graphically the results of parametric analysis on ventilation rate of atria in relation to the openings area and the height of the atria. In the General advice part of ISOLDE the use of natural ventilation is recommended in relation to values of wind velocity and building height. The information related to natural ventilation appear in different screens. 159

Advantages

• Can improve thermal comfort of occupants • When air temperature is lower than internal air the structure of the building can be cooled without any energy consumption Disadvantages: • Wind induced in the rooms may create discomfort in working places Side effect:

• Orientation according to wind direction for best ventilation may be conflicting with orientation for best solar protection • Wind ventilation is severly altered by earth sheltering, city compactness strategies, use of vegetation for shadowing Fig. 4 presents general advices on the selected cooling principle. For specific situation also quantitative informations will be provided. For example atria are more and more used in energy efficient commercial buildings. Information will be provided from momitored buildings and simulations that are carried in the Task XI. We will here present for this specific component some indications taken from literature. Natural ventilation can be generated by a solar driven buoyancy effect in all sizeable atria. The air movement is related to the height of the atria, the openings area, the climatic data. In fig. 5 the air change rate and vertical air velocities have been calculated for different configurations of an atrium enclosed by a building of masonry construction during a day with an average temperature of 20 °C and a maximum temperature of 25 °C (Owens, 1987). 2.2. Forced ventilation

This option is suggested when the mean daily temperature swing during summer months is higher than 12 °C and the mean value of the maximum daily temperature is lower than 30 °C. ♦ο τ 10 m

O 30 ­

S, S

20 m

g> 20 + 30 m

Õ

5

IO

y. 160

opening

15

S

S Ρ

Fig. 5 Natural ventilation in atria due to buoyancy effect

In the following screen a set of suggestions to be kept in mind when using this strategy is proposed. The building needs good thermal inertia (equivalent to more than 400 kg concrete per sqm of floor area) Large Surface area of the "mass" elements (about 4 times the floor area) High outside insulation Ceilings or floors with embedded channels recommended Caution with daytime natural ventilation Define appropriate control strategies for ventilator fans Fig. 6 General recommendations for the use offorced night ventilation A second screen presents some quantitative data on nocturnal forced ventilation for different climates (Givoni, 1988). Climate Hot-Humid Hoi-Arid Temperate

Ta(°C) 24 21 18

Tm (max)-Tm(min) 3 7 9

QIA (Wlmq) 14.8 34.5 44.3

COP 4.4 10.3 13 2

Ta average summer night ambient temperate (Jun, Jul, Aug ) in N e * Orleans, LA. Phoenix, A.Z and New York Tm (max) (min) average maximum (minimum) - temperature Q/A rate of cold storage vs area COP coefficient of performance (ratio cold storage vs energy for the fan)

Tab. 2 Efficiency of forced natural ventilation in single storey buildings calculated for different climates 2.2.1 Use of hollow core ceilings As in the case of atria informations taken from buildings monitored in the Task XI will be presented. In the following figures the stratgegy is described and some experimental data measured during the last week of July 1989 in an office building located in Giarre, near Catania, Siciliy are shown (Silvestrini, 1989 b). Parametric analysis using computer simulation models will give more general informations on this strategy.

Fig. 7 Night ventilation cooling through hollow core ceiling in a Barra-Costantini passive component 161

Ext Temperature

Cooled room

26

Fig. 8 External air temperatures measured in darre during the last week of July 1989

27

28

29

30

31

1

Fig. 9 Temperature values of the cooled and the reference ceiling monitored in darre

3. Conclusions The work in progress on the cooling part of an expert system, called ISOLDE, to be used as an energy consultant has been described. The knowledge bases and rules will include all the informations that will be elaborated in the IEA Solar Heating and Cooling Task 11 and of more general indications found in literature.

Note: The working group on Expert Systems of IEA Task XI is composed by Ove Morck (coordinator), Ida Bryn, Nicolas Morel, Rolf Strieker, Giovanni Silvestrini

4. References Baker N. (1986), "Passive cooling evaluation method - PACE", Proc. 5th PLEA Conf., Pecs, Hungary Bauman F. (1988), Ernest D., Arens E., "The effects of surrounding buildings on wind pressure distributions and natural ventilation in long building rows", Ashrae Transactions Baer S. (1983), "Raising the 'open U' value by passive means", Proc. 8th Passive solar conference, ASES, Santa Fe Dick J. (1950), "The fundamentals of natural ventilation of houses", Journal of the Institution of Heating and Ventilating Engineers, june 1950 D.W. Etheridge (1988), R J . Stanway, "A parametric study of ventilation as a basis for design", Building and Environment, Vol. 23, No. 2, pp. 81-93,1988 162

Evans Β., (1979), "The study and use of natural air flow in buildings", 3th Passive National Conference, ASES, San Jose' Feustel H., Sherman M., (1989), "A simplified model for predicting air flow in multizone structures", Energy and B uildings, Vol 13, n. 3, Elsevier Sequoia ed. Liddament M. (1988), "The calculation of wind effect on ventilation", Ashrae Transac­ tions n. 13 A.G. Loudon (1968), "Summertime temperatures in buildings", B rs Current Papers CP47/68 Mathews E. (1986), "Thermal analysis of naturally ventilated buildings", Building and Environment, Vol. 21, No. 21 Mathews E., Richards P. (1989), "A tool for predicting hourly air temperatures and sensible energy loads in buildings at sketch design stage", Energy and buildings n. 14 Mc Farland R., (1988), Lazarus G., "Monthly auxiliary cooling estimation for residential buildings", LA­11394­MS, Preprint, Los Alamos Niles P. (1986), "Equation to Predict the Temperatures Inside of a night ventilation cooled building on a design day", 11th National Passive Conference, ASES, B oulder Owens P. (1987), "Natural ventilation of atria", European Conference on Architecture, Munich, 6­10 april 1987 Silvestrini G., Alesandro S., (1989), "Algorithms and design tools for natural ventilation cooling of buildings", Congreso Internacional Energia Ambiente e Innovación Tecnologi­ ca, Caracas Silvestrini G, (1989, b), "Advanced case study ­ Sogeco office building; capabilities of simulation models to evaluate cooling strategies", internal report Task XI IEA expert meeting, Oakland

163

Natural Cooling Techniques S.A. Domínguez Dept. de Ingeniería Energética y Mecánica de Fluidos - Escuela Superior de Ingenieros Industriales Avda. Reina Mercedes s/n - 41012 Seville - Spain

ABSTRAGT. This paper is not intended as an exhaustive account of all the available natural cooling techniques, including an interminable list of references, the advantages and disadvantages of each technique, and their field of application. Rather, it is intended to present the perspectives of future research and the topics t h a t should be covered within the frame of a European concerted action. Thus, it will concentrate on the main obstacles and faults t h a t we feel exist in the current research, and subsequently on the measures t h a t should be taken to correct this. 1. CLASSIFICATION O F NATURAL COOLING T E C H N I Q U E S . The opinions of diverse authors don't agree regarding the best way to classify natural cooling techniques. Thus, these classifications can be made according to: -

Nature of the heat sinks (ambient air, sky, etc.) Heat and mass transfer phenomena involved (convection, evaporation, etc.) Storage period (daily, weekly, seasonal) Storage material (water, rock, etc.) Type of application (hot/dry climate, hot/humid climate)

The diversity of this classification yields to a certain confusion between the causes and effects, with the result t h a t identical strategies are often considered from very different points of view. For example, the ground in earth-coupled structures, considered a heat sink by some authors, is treated as seasonal storage material by others. These questions have a conceptual interest, but are irrelevant to the objectives of this presentation. For methodological reasons, natural cooling techniques will be classified depending on how they influence the thermal behavior of the building. We will distinguish between: - Direct cooling systems, where the heat sink promotes a direct cooling action on the building structure, a n d / o r the interior air. 165

- Indirect cooling systems, where the system first cools a specific amount of air, which is then discharged (with or without intermediate storage) to the building. This classification can be further clarified by the following examples: - The ground in contact with the building structure is a direct cooling system, but the ground used as a means to cool the air blowing through an underground duct is an indirect cooling system. - Roof ponds, roof sprays, or roof water films are examples of direct cooling systems, but the water used in mechanical evaporation cooling units, wind towers, or cool-towers (as the one proposed in the Appendix), are considered indirect cooling systems. - Typologies used to promote cross ventilation to produce comfort and the nocturnal ventilation strategies used to cool the building structure are direct cooling systems. Obviously, there are cases where certain systems will not exactly fit this classification. These systems can be considered as mixed cooling systems and will not modify the conclusions of this paper. 2. P R O B L E M S C O N N E C T E D T O C U R R E N T RESEARCH ON DIR E C T COOLING SYSTEMS. Independently of the fact t h a t direct cooling systems may be elements inherent to the building structure (roof pond, underground walls, etc.) or not (ventilation, strategies, etc.), a characteristic common to all of them is the strong coupling between their thermal behavior and t h a t of the building. For instance, in earth-contact structures, the building modifies the ground's temperature field; hence evaluation of their thermal performance requires realistic formulation of the boundary conditions, especially if surface treatments are accomplished to improve the ground cooling potential (shadow on adjacent spaces, irrigation, etc.). This degree of detail implies the use of two-dimensional or three-dimensional transient conduction. However, in most computer simulation codes used for buildings' energy analyses, thermal conduction is considered one-dimensional. Thus, the only way they treat conduction through the ground is via an "assumed" ground temperature. Conceptually, this situation is very similar in other direct cooling systems. Evaluation (thermal performance) of a certain component usually require a level of complexity t h a t infers the use of certain modelling technique. If the computer code available to simulate the building uses a different modelling technique, the component cannot be "integrated" into the code, or, at most, this integration will be made based on simplified assumptions t h a t disturb the evaluation results. With regards to ventilation strategies, the same difficulties are encountered. Many of the existing computer codes are unable to properly manage 166

multizone air flow exchanges, especially if the building includes typologies oriented to promote ventilation (e.g., solar chimneys). The problem here is the impossibility of modelling the thermal and aeraulie couplings. A demostration of the magnitude of errors that occur when a ventilation strategy is inadequately modelled is show in Figure 1 the results of an interprogram comparison exercise. Oases 100 to 210 represent changes in design parameters of a reference cell - e.g., inertia, absorptivity of the walls, window area, etc.; cases 220 and 230 represent the results for low mass and high mass cells, respectively, when nocturnal ventilation has been included. As can be seen, the differences between the programs -acceptable for cases 100 to 210 - are considerably increased when considering ventilation. In sum, the major problem encountered in the evaluation of direct cooling systems is that the strong coupling existent, makes them inviable for realistic treatment in most computer codes. Consequently, we cannot properly analyze the true effect of a component, system, or strategy on the building's performance.

IEA COOLBOX ANNUAL COOLING MEMPHIS KWH (Thousands)

100

110

120

130

140

150

160

170

180

190

200

210

220

230

CASES SERIRnS

Ë 2 3 DOE2

S3PAS

E S 3 TASE

ESP

FEASIBILITY STUDY 275/90

Fig.l

Interprogram

comparison

exercise

3. PROBLEMS CONNECTED TO CURRENT RESEARCH ON INDIRECT COOLING SYSTEMS. Scientific literature reports hundreds of different manners to cool the air using natural means. 167

An inspection of technical papers reveals t h a t most of them deal with research t h a t includes monitoring, superficial analysis of the results usually in terms of efficiency, and in some cases, modelling based on more or less sophisticated correlations. Apart from a few exceptions, the design of the system has been made based on intuition and without the use of even the most rudimentary calculation methods. After the experiment (usually poorly monitorized), there is a lack of in-depth analysis directed to explain the reasons for the results or the possible design failures. Finally, there is usually a lack of modelling and validation activities; consequently, a sensitivity analysis is out of the question, and the opportunity to determine the following is lost: - Which design parameters are relevant? - How should these design parameters be modified to improve the design? - W h a t is the optimum range of variation of each design parameter according to different climatic conditions? Consequently, the results of research are very limited, from which we obtain little more than a demonstration plant. In light of this situation, and taking into account the large degree of variation t h a t each cooling technique allows, it has not yet been possible to obtain a sound basis for systematic extrapolation of the results, and the elaboration of the subsequent design guidelines.

4. PROPOSALS FOR FUTURE RESEARCH. 4.1. Direct Cooling Techniques In my opinion, the first action to be taken in research to promote the understanding, enhancement, and use of direct cooling techniques, is to develop general coupling methodologies. These would allow the integration in computer codes of algorithms pertaining to components or strategies, with no restrictions on the modelling technique or the degree of complexity. In addition, further research must be performed to improve the characterization of some direct cooling techniques and strategies (e.g., ground coupled structures and multizone air flow). 4.2. Indirect Cooling Techniques To obtain a sound background t h a t enables the elaboration of design guidelines for indirect cooling techniques, it would be useful to promote specific studies which should include: - Modelling of the heat and mass transfer phenomena involved,explicitly including all the design parameters 168

- Fully monitorized experiments - Rigorous validation - Sensitivity analysis The Appendix includes an example t h a t illustrates the proposed methodology. An indirect cooling technique is presented, intended to use for climate conditioning of open spaces in the Expo'92 Universal Exhibition. However, these type of actions are unsuitable for a multinational research program, for the following reasons: - There are too many systems. - Different systems are usually based on different physical principles. - Each country or region would be interested only in a limited number of actions, depending on their own climate peculiarities and natural resources - The study of each system will not require the conjunction of researchers from different groups. Consequently, it will be more convenient to perform these studies on a national, or even regional, level. A European concerted project could link these efforts through common methodology of work and documentation. Concerning the integration of the indirect cooling techniques used in building performance, we highlight a relevant topic t h a t could be included in a Community concerted project. This topic has been partially mentioned for direct cooling techniques and consists of the development of formulations to characterize the coupling of thermal models, air flow models, and wat e r / v a p o r transport models. The inclusion of water/vapor transport models is of great interest from a comfort point of view, to evaluate the effect of indirect cooling systems based on adiabatic saturation processes.

169

APPENDIX: COOLING-TOWER PERFORMANCE This Appendix is a short presentation to illustrate the possibilities of research carried out according to the methodology previously cited. The system (Figure 2) consists of a transparent P V C tower, with nozzles t h a t create an artifical fog by injecting water at high pressure through minute orifices. Since the water droplets are very small, their evaporation is almost instantaneous. This evaporation cools the air blown by a fan placed at the t o p of the tower. The research carried out on this tower included the following: - Modelling of the evaporation of a single water droplet in the air - Characterization of the droplet size distribution produced by the fog generators (micronizers). - Modelling the simultaneous heat and mass transfer phenomena occurring when air flow and fog (droplets of different sizes) come into contact in the tower. - Design of the experiment (select the different variables). - Experimentation (summer 1989). - Qualitative analysis of results (Figures 3 and 4). - Validation (Figure 5). - Sensitivity analysis, in which the following items were examined: - Influence of the air flowrate (m'/A.). - Influence of the quality of micronizers in terms of their VMD (Volume Median Diameter): better quality infers lower V M D . - Influence of the water flowrate {l/mx of air). - Influence of the micronizers location within the tower. - Influence of the tower's height (distance between the base and the micronizers). - Influence of ey+erior conditions (ambient temperature and relative humidity). Figures 6 and 7 show, respectively, the outlet temperature and the percentage of evaporated water as a function of the tower height and the VMD of the micronizers. The sensitivity analysis provided us with the knowledge of the influence of the relevant design parameters, through which we were able to establish design guidelines for this kind of systems. For a new application, we are now able to specify all the design variables and the control strategy to maximize the cooling capacity, while avoiding people from getting wet because of non-evaporated water droplets. 170

Fig.2.

The c o o l i n g

DIA INICIAL:

tower

Χ IB

AGOSTO B 9

T

f:­i.

SA: : D A

TCRÍÍF

TEM. ENTRADA TORRE O)

m

1

(

1

1

1

1

1

1

m

\

/ m

/ O

O"

}

— m ■




Vul\

ι

íT~^V

_

^-

^

.

.

ν.

>

-

^

.,\.

.

:

\

^ - " V

1\

ν

^ s

^

V^

s

Grbùnd. ^ - -Ν·

'..

. S s ^ ' 'V.s

—».

.

—»*

S

'

■ '. S

s

^

^

s

V .

s

\

ν

■ S

S

■ V -

.

ν

'

' ^ ·

.
€% ?C%

U.5%

17.5%

44.5%

20%

80%

4'7%

SIX

ΛΓΧ

1G%

30%

66%

34%

β'Λ

13%

1G%

38%

6?»

4­r%

23%

33%

3β%

6 3

5β%

7%

35%

49%

51%

31%

41%

28%

TABLE I S e l e c t e d Solar Passive Buildings in Mediterranean a r e a f r o m p u b l i s h e d P r oj e c t M o n i t o r S c h e d u l e s . We c a n o b s e r v e t h a n t h e 31% a t l e a s t summer t r a c k ­ h a s n o t b e e n c a r r i e d o u t . Most of t h e b u i l d i n g s monitorized present overheating problems during summer ( 7 0 % ) , when t h e s i t e i s c l e a r l y M e d i t e r r a ­ n e a n . No summer p a s s i v e s y s t e m s a r e u s e d i n t h e s e l e c t e d b u i l d i n g s , p r a c t i c a l l y , they only e x i s t for the heating season.

186

τσ%

27%

2.­ EXPERIMENTAL

DIFFICULTIES

Different problems have been found out in building monito­ ring during summer periods. We specially mention those related with the use of simple methods of data analysis, i.e., degree days based energy evaluation methods. a. The main problem is the ACH measurements. These measure­ ments can be carried out by means of SF release system / 2 / , composed by a small SF cilinder, a regulator and a solenoid valve (Figure 1 ) . This system also requires a cromatographic oven, and electron capture detector, a microcomputer, an inter­ face set and a sophisticated calibration system. Obviously in full scale monitorings, in the same way used in the Project Monitor development, such an experimental studies are not ava­ ilable because their economical costs, personnel expenses and also users tolerance in inhabited buildings. b. In high solar irradiation climates the ability to ge­ nerate convective loops with heat distribution into the buil­ ding is an important aspect to be considered. These energy flows affects the global building efficiency in winter and summer periods. Internal distribution is a phenomena that can be taken into account and it must be measured in a specific way, to evaluate the building dissipative capability (Figure 2). c. The use of remote sensing techniques (like satellite image analysis or local infrared thermography) give us a com­ plementary characterization of several microclimatic aspects, and also convective and radiative details. d. The other hand we cannot forget the evaluation of na­ tural cooling by latent heat transformation through humidity measurements, and also intensity and direction of wind.

OUTSIDE

DATA

Ξ1

^

ARGON GAS

ACQUISITIQN

FIGURE Schematic diagram of the SF system, used for measurement of air flow and air change in houses. 187

FIGURE 2 The graphical representation of the North/Sud temperature in a house is the more simple test to study the heat flows in between. In photo, the example of Bègues Experimental Field of 20 dwellings (Barcelona).

It is obvious, that from an experimental point of view it represents a qualitative change when comparing with finished Project Monitor, in reference with economical costs as much as personnel requirements. But it is the logical continuation of experimental research in the field of Bioclimatic Architecture. In this way, we can take advantage of the accumulated knowledge / 3 / .

3.- PROBLEMS OF THE SIMPLE METHODS In many cases monitoring during heating season has been analysed by the substractive method. In this method solar contribution of passive systems can be deduced from building heat losses, the internal gains and auxiliary energy used by the expression: Q s = Q ref_ - 0 aux - Q int Where solar contribution includes not only energy gains obtained by the use of passive solar systems but also energetic contribution of thermal inertia of the building. In this way, building monitoring with high inertia(such as semiburied houses) during mild periods can show negative solar 188

c o n t r i b u t i o n in the sense of the above d e f i n i t i o n , due to p e r f o r m a n c e of great a m m o u n t of thermal m a s s .

the

An a t t e m p t to adapt this simple c a l c u l u s model for summer building e v a l u a t i o n will be to r e f o r m u l a t e the above e x p r e s ­ sion to : =

( v0 v

ref

+ 0 Vef

) ­ Q. aux

- 0int

Where ref

can be u n d e r s t o o d as the usual b u i l d i n g heat losses (that can be m i n o r to z e r o ,

when θ.i - 9 o < 0 ) . ref

is the heat e v a c u a t i o n by m e a n s of n a t u r a l cooling s y s t e m s : v e n t i l a t i o n , e v a p o r a t i v e or r a d i a t i v e c o o l i n g , e t c .

In any case b u i l d i n g heat losses v a l u e will be c a l c u l a ­ ted through B uilding Load C o e f f i c i e n t . In summer p e r i o d s this c o e f f i c i e n t has a great degree of u n c e r t a i n i n i t y , mainly related to v e n t i l a t i o n which has to e f f e c t s : ­ The t u r b u l e n t regime of air close to the e n c l o s u r e surfaces i n c r e a s e s energetic exchange. ­ A . C . H . values can be the g r e a t c o n t r i b u t i o n to the B . L . C , v a l u e , and have important e x p e r i m e n t a l d i f f i c u l t i e s in m e a s u r e m e n t and variability. In the other hand the e v a l u a t i o n of internal g a i n s summer is affected by a d i f f e r e n t building u s e .

during

In order to evaluate energetic e f f i c i e n c y of p a s s i v e c o ­ oling systems the use of the above e x p r e s s i o n doesn't to be a p p r o p r i a t e to deduce Q _ because of the above m e n t i o n e d ΓGI

d i f f i c u l t i e s to d e t e r m i n e the e f f e c t i v e B LC l o s s e s , and on the other hand the lack of an a l t e r n a t i v e way to deduce or measure a rough value of Q has no s i g n i f i c a n c e in p a s s i v e cooling s y s t e m s such as passive h e a t i n g s y s t e m s , ( F i g u r e 3 ) . To try to solve this in a simple way we can try out the r e g r e s s i v e a n a l y s i s or the i d e n t i f i c a t i o n m e t h o d s with some adaptations.

Heat Flux to e o l i liaat riu* to aneltered

lt«at Flua to e s t . « a l l s

Neat Flux to c e l l i n g Day ( J u l / 19881

FIGURE

3

Positive and negative heat flux in a semi sheltered building, du­ ring summer.( See Llavaneres Se­ miburied House in brochure 37 of the Project Monitor). Reference to Q ^ and Q ref ref 189

4.­ FEASIB ILITY OF USE THE IDENTIFICATION METHOD A suitable method to analyse a real building monitoring in the summer period can be the use of identification tech­ niques to deduce thermal parameter values that cannot be calculated in a difieren t way. Subi ntervals of about 15 days can serve as initial per iods for a previous study to estimate their variation range, ρ rovided of an appropiate variability However, we can remark that the of external solicitation s /A/. use of well documented m ethods in c ell tests, will obviously give us a lower degree o f accuracy in real cases, but it will allow us to overcome a d ifficult im passe. Parameters subjected to the above mentioned previous stu­ dy are: — Thermal resistance of interior surfaces: R as — Air Renovation Index: A C . H . — Thermal resistance and c apacity R and C of g g ground exchange. — Effective Solar Area: A g¡apacity, R and C,of — Thermal resistance and c the rest of the building And the associated model, with their connectivity can be the presented in Figure 4.

A.C.H.

e,

θ

FIGURE

4

Proposed energy balance model with only mention of parameters to be identified. 190

The experimental inputs in this case will be: ­ Solar Irradiation on vertical South facade and other projected surfaces. ­ Auxiliary Energy consumption. ­ Internal Energy Gains. ­ Time—temperature functions. ­ Mean relative Humidity. And the set of equations describing the model are

C ·θ g ι •&

θι - e g

ôi - 0 «i -

0 + 0 W *LAT AUX

G i ­

G i ­ ^ 2 + -

W

+ I A INT

R

s

0 1 —

+ R

+ 0

O

w

Λ +

ACH'V· (tf. ­

+

R

as

Λ θο ) " C. Ρ

R g

Where Rw = Thermal resistance of external walls without thermal resistance of interior surfaces. INT Q

­ Internal Energy Gains = Auxiliar Energy consumption.

AUÄ

I

= Solar Irradiation on vertical South surface.

V

= Internal volume.

This set of equations can be solved, aproximately, by numerical methods. In order to establish a comparative study of summer confort conscious buildings in different locations, in a similar way than Project Monitor, a solving tool should be developed to be used for all research groups involved in this field. An additional complexity can be taken into account: the estimated parameter values calculated in a selected subperiod must be projected on time in order to evaluate the confidence interval where this set of parameters can be extrapolated. The simple stationary methods used in heating periods doesn't seem to be appropiate for summer behaviour analysis. As an alternative we propose to obviate the time variable, but with a previous identification of average parameters deduced from time­dependence of experimental variables.

191

5.- CONCLUSSIONS Experimental support is needed to study thermal behaviour of buildings during surplus energetic periods. However thermoconstructive and climatic variables control cannot be carried out in such a simple way as in heating season. Some difficulties of experimental nature, of energetic evaluation and interpretation of final results has been reported. It is obvious that the progress on research programs like Project Monitor, specially focussed on Passive Cooling Systems, need of new sophisticated experimental and economical resources However, we can diminish them by using data analysis methods that allow us to estimate thermal parameters without any direct measurement (i.e. A C H ) , maybe through identification method adaptation with a sufficient approximation degree. An interesting goal will be to create a research group in the E.E.C, scope to generate a practical tool for building monitoring in summer periods.

REFERENCES « / 1 / Overheating in Bioclimatic Architecture in Mediterranean Climates, J.A.Cusido, J.J.Escobar, J.Esteve, J. Jorge, A.Mitjà, J.Puigdomènech. Science and Technology at the Service of Architecture, 4-8 Dec.,1989 (Paris). / 2 / Measurement of Air Flow Between the Floors of Houses Using a Portable SF System, S.B.Riffat, M.Eid. Energy and Buildings, 12, pag. 67—75, 1988. 13/

Calculation Methods of Energy Consumption for Summer Conditioning of Buildings, M.Cucumo, V.Marinelli, D. Kaliaksos, Proc. of the Evolution of External Perimetral Components in Bioclimatic Architecture, ISES, 5-6 April,1990 (Milan).

/ 4 / Characterization of the Variation of the Solar Aperture of a Direct Gain Equipped with an Automatic on—off Controlled Shading Device, P.Achard, M.Tantôt. Workshop on Passive Solar Testing in Mediterranean Climates, J.R.C., 24-26 June, 1987 (Ispra).

192

Session 8 Additional Heating and Cooling

Auxiliary Environmental Control in Passively Cooled Buildings E. Maldonado Dept. of Mechanical Engineering - University of Porto

R. Bragas 4099 Porto Codex - Portugal

ABSTRACT. The need for auxiliary heating and cooling systems in European buildings is established on the basis of building physics and climatic conditions, emphasizing that cooling systems may not be needed in most regions if there are no large internal gains and the building envelope is well designed, through the use of bioclimatic design principles. Occupant attitudes and the consequences upon indoor environmental quality are also compared for the cases of actively and naturally controlled buildings. Ways to optimize the energy and functional efficiencies of mechanical cooling systems are then discussed, including the cooling-producing equipment, distribution networks and control strategies. Finally, the major difficulties that now exist in modelling, sizing and designing HVAC systems are also presented, pointing out some topics for future research and standardization.

1.

INTRODUCTION

Auxiliary systems for indoor environmental control, from the simplest fireplace to the most sophisticated heating, ventilating and airconditioning computer-controlled system, are present in the vast majority of the building stock, whatever the type of, or the use for the buildings. Their need is seldom questioned, but the best type or size for a particular application is often debatable, especially when energy efficiency considerations come into play. This debate becomes more important in buildings where bioclimatic design procedures are adopted, as their dynamic thermal behaviour is usually more sensitive to interactions with any active control systems that might be functioning within it, particularly, once again, if it desired to take the most advantage of passive solar. In this text, prior to discussing the auxiliary systems themselves and the critical issues that can now be raised in terms of their design and energy efficiency, the need for such systems will be critically evaluated or, better yet, bound by a set of objective parameters. Cooling will naturally receive most of the attention.

2.

THE NEED FOR ADDITIONAL HEATING AND COOLING

Auxiliary energy inputs to an the indoor environment, i.e., within a range around desired needed at a certain moment is

indoor space serve the purpose of keeping temperature and, in some cases, humidity, setpoints. The amount of auxiliary energy the balance of all the thermal exchanges 195

from the building envelope to the indoor air and the magnitude available indoor heat sources, as shown in Fig.l and eqn.(l): Qaux ­

(1)

Qge + Qgi + Q;

Q gaina,

of the

envelope

gaina, ventilation

(V) Fig.l ­ Instantaneous Space Heat B alance

The two envelope terms (Q and Q ) can be either positive (gains) or negative (losses). If the gains exceed the losses, cooling is required to maintain a constant indoor temperature T­. Conversely, if losses exceed the gains, Q aux becomes negative and heating is therefore required. All these energy terms are dynamic in nature, due to outdoor climatic influences and to interactions with thermal storage in the building mass, as will be discussed later on with more detail. B ut it is useful to also analyze the heat balance in terms of integrated means over a sufficiently long period of time, as this will allow a meaningfull insight towards establishing the need of auxiliary energy systems. Indeed, the ultimate aim of bioclimatic design is the "zero energy" house, i.e., a house where no auxiliary energy would be needed. Then, realizing that V / C p (Ta­T¡)

*

/ ''/ /

Κ

/ o>a O

y

0Λ°

l·/ ' /

3.2 -Ζ 0

τ?

s

,¿# ¿

2.0 _Ε

* < ^

0.0»

/

Ο.*

^

y/y

1.6 _Ξ

Í



0.04

V,

1.2 _=

%%.

0.8 _= 0.4 _Ζ

11 1 o

ÍN.

o Γ*

β

ΓΝ

I

«>

«o

to

1

iri

iri

»ri

tri

V

V

V

V

ri

ri

η

ri

pj

w

1 ! 1 1 1 1 1 DIRECT GAIN SYSTEMS

»­J

c*

1

^

­^

1 1 ;

RATIO MASS / ÕLA 2ZD ARM ­ 9 : s u o i : / Douei£

jcmiNC STjiai DC 7

11

| 1 | 1 | 1 | 1 ¡ 1 | 1 | 1 | 1 ¡ 1 | 1 | 1

:

mem INSULA OON

: THERUAL CA PA CITY

A KSSI NCE

oanU

11c

Ί

o

4 13 2 19 20 14 11 12 5 18 15

o

ι /

24 26

17

o

O

—0.10

—0.20

s

25 27 10 6­23 22 3 2B oc. 1 ­ 7 5­16­21