Technical Note TN 4/99 4 AIR DISTRIBUTION 4 IN NATURALLY VENTILATED OFFICES Jw PJ Jackman JDETR emronaaer Ma ree frreThe Building Services Research and Information Association Old Bracknell Lane West, Bracknell, Berkshire RG12 7AH UK Tol: +44 (0) 1344 426511 Fax: +44 (0) 1344 487575 e-mail: bsria@bstia.co.uk web: www.bsria.co.uk Technical Note TN 4/99 AIR DISTRIBUTION IN NATURALLY VENTILATED OFFICES P J Jackman BRAAll rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted in any form or by any means electronic, mechanical, photocopying, recording or otherwise without prior written permission of the publishers. ISBN 0 86022 5224 Printed by Oakdale Printing Co, BSRIA 77880 February 1999ACKNOWLEDGEMENTS ‘This work was part-funded by the Department of the Environment, ‘Transport and the Regions (DETR), under the Partners in Innovation scheme, BSRIA acknowledges the financial support of the DETR and would like to thank the following sponsors for their contribution which has led to the production of this Technical Note, wrnowwen S DETR ‘Department of the Environment, Transport and the Regions. BBC Environmental Services Flomerics Ltd Greenwood Airvac HGa Consulting Engineers Rybka Battle UK Ltd Schiico International The research project was undertaken under the guidance of a project steering group. The steering group contributors were: Mr J Agombar (BBC Environmental Services) Dr R Cohen (HGa Consulting Engineers) Mr M Gardner (Greenwood Airvac) Mr C Matson (Rybka Battle UK Ltd) Mr M Seymour (Flomerics Ltd) Mr J Stamp (Schiico International) Mr M White (Building Research Establishment) Further thanks is given to Dr R Cohen, who acted as a Project Assessor. ‘This publication is issued with the agreement of the DETR and every ‘opportunity has been taken to incorporate the views of the steering, ‘group but final editorial control of this document rests with BSRIA. OBSRIA TN 4709 ‘Air Distbution in Naturally Ventilated OffcosCONTENTS 1 INTRODUCTION 2D IGN GUIDELINES 3Rl RCH RESULTS 3.1 Ventilation rates . 3.2 Internal temperatures 3.3 Thermal comfort. 3.4 Ventilation air flow pattern 3.5 Depth of ventilation. 3.6 Window type - 3.7 Variation between floors 3.8 Required window opening 4THE STUDY 5 REFERENCE LIST OF TABLES ‘Table 1 Physical and Thermal Characteristics of the offices LIST OF FIGURES Figure 1 ‘Typical variation of ventilation rate with window opening Figure 2 ‘Typical variation of room air temperature with window opening, Figure 3. Typical variation of perceived discomfort with window opening Figure 4 Typical variation of local ventilation rate with distance from window Figure 5 Typical variation of room air temperature with distance from window ... Figure 6 Ventilation air flow pattern - 6m office Figure 7 Ventilation air flow pattern - 12 m office Figure 8 Comparative effectiveness of window types. Figure 9 Window types. Figure 10 Ventilation rates per floor with window opening ratio of 0.077 and room heat load of 14 Wim? Figure 11 Average ar temperature per floor with window opening ratio of 0.077 and room heat load of 14 Wim? Figure 12 Percentage dissatisfied per floor with window opening ratio of 0.077 and room heat load of 14 Win Figure 13 Minimum open are ratio required for centre-pivoted windows Figure 14 Minimum open are ratio required for top-hung windows Figure 15 Elevation and plan of simulated office. 14 u 1B 13 15 Air Distibution in Naturally Ventlated Offices (BBSRIATN 409,4 INTRODUCTION Natural ventilation is widely used to provide a supply of outdoor air into buildings. In winter-time the main requirement is to provide sufficient air exchange to dilute and remove aitbome pollutants such as carbon dioxide and body odour emitted by people and any impurities released into the air from other sources. Maintaining acceptable moisture content of the indoor air may also be an important criterion. In summer-time there is an additional requirement to ventilate sufficiently to limit the increase of indoor temperature to an acceptably comfortable level This publication focuses on the summer-time condition and presents results from a BSRIA study [1], part-funded by the Department of nviconment, Transport and the Regions (ref: 38/6/96) under the Partners in Technology scheme (now Partners in Innovation). The ‘main purpose of this study was to provide guidelines on the factors required to achieve successful natural ventilation in offices in summer by the use of openable windows. ‘A combination of physical and computer modelling was used to ‘examine the limiting conditions for natural ventilation through one external facade to be effective in providing a comfortable and healthy indoor environment on a still, warm summer day: ‘The first section of this document contains the design guidelines, ‘Subsequent sections summarise the research and the findings from which the guidelines were compiled, GBSRIA TN 4190 Air Distibution in Naturaly Ventilated Offices2 DESIGN GUIDELINES ‘Single-sided ventilation is a viable means of supplying outdoor air to offices. The study showed that, with an open window area of 1/20" of the floor area (the value specified in Approved Document Part F of the Building Regulations [2]) for “rapid ventilation” in non-domestic occupied buildings), the ventilation flow rate is more than adequate to meet the specified requirements of 8 litres/s per person and 1.3 litres/s per m’ of floor area [3] over the range of conditions examined, Average indoor temperatures reduce with increasing window opening as a consequence of increased ventilation rates. Thermal comfort, rather than ventilation rate, is the limiting factor in summer-time so the viability of single-sided ventilation is restricted to situations where the external and internal heat gains are not excessive. The study assumed full solar shading (ie no direct solar radiation). ‘The results indicate that, with an internal heat load greater than 10 Wim", window opening areas greater than 1/20" of the floor area (>0.05 m’/m’) are required to maintain comfortable conditions. ‘The pattern of ventilation air flow is such that cooler air from ‘outdoors moves across the office at low level, is drawn upwards into thermal plumes above people and heat generating equipment, and returns to the window end of the office at high level. Local ventilation rates and air temperatures do not vary significantly with distance from the windows Single-sided ventilation can be applied in offices approaching 12 1m deep, if windows with sufficiently large openable window areas can be provided in the external wall. ‘The previbusly defined limit, of 10 mis an acceptable criterion [4] Ventilation is most effective with window openings at both high and low level. The sash windows are good, but the centrally pivoted and top-hung ones were even better because of the potential for larger open areas. ‘The rate of overall ventilation, as expected, increases with the open area of the window, with improvements found right up to the maximum of 3.5 m’ (0.116 m*/m’ of floor area in the 6m office). @RSRIA TN 409 ‘Air Disioution in Naturally Ventlated Offices3 RESEARCH RESULTS 3.4 VENTILATION RATES 3.2 INTERNAL ‘TEMPERATURES 3.3 THERMAL COMFORT As expected, the overall ventilation rate increases with the open area of the window, irrespective of the window type, Figure 1 shows a typical ‘example that illustrates linear relationships between ventilation rate and open area for three room heat loads. In the chosen summer time cease, the ventilation is thermally driven and Figure | also shows the increase in ventilation rate with increase in internal load. The minimum acceptable ventilation rates specified in BS 5925(3] are 8 litres/s per person or 1.3 litres/s per m’ of floor area. For the offices used in this study, these ventilation rates are equivalent to 1.0 and 1.7 air changes per hour. A review of the research results shows that these levels of ventilation were always exceeded when the window ‘opening was at least 1/20" of the floor area. This window opening area is specified as a requirement in Approved Document F of the Building Regulations [2] for “rapid ventilation” in non-domestic ‘occupied buildings that rely on natural ventilation. Figure 2 shows the variation in the average room air temperature with increasing window opening. Tt is based on the same conditions as Figure 1. The relationship is such that room air temperatures are reduced with higher window openings, this being the effect of higher ventilation rates. The impact of higher heat loads on internal temperatures is also clearly seen. Under low heat loads and with the better window types, average indoor temperatures are less than 2°C above that outdoors Thermal comfort depends largely on a combination of air movement and temperature and these parameters can be combined to predict the number of occupants likely to uncomfortably warm or cool, One such index is PPD (predicted percentage dissatisfied) and this has been used to assess the conditions under which thermal comfort may be achieved. As shown in Figure 3, a typical relationship between PPD and ‘window opening shows the benefits of increased open areas, ‘This is the consequence of the higher ventilation rates and reduced room air temperatures Air Distribution in Naturally Ventiated Oficas (BSRIA TN 419914 - ow = = 2 10 Figure 1 £8 : — Typical variation of a6 oo - ventilation rate with 5 A window opening $4 2 a . oO 1 2 3 4 ‘open area (m?) ¢ 32 — 2 * s 3 - wy Figure 2 BOT omnes NE Typical variation of SO 29 1 coa (wim?) ~ room air temperature =p 28 with window opening 88 2 J 3 26 § 25+ a 24 — o 1 2 a 3 4 open area (m 90 - 2 80 - 70 +— Figure 3 23 60 Typical variation of 3 50 a perceived discomfort a3 eS room heat with window opening Sq 4 AS load (wim?) $5 30+— a 5 20 > 2 iol o- — 0 1 2 3 4 open area (m*) @ASRIATN 4199 ‘Air isibution in Naturally Ventilated OfficesSTS 3.4 VENTILATION AIR FLOW PATTERN 3.5 DEPTHOF VENTILATION Figure 4 ‘Typical variation of local ventitation rate with cistance from window Figure 5 Typical variation of room air temperature with distance from window air change rate (h”) ‘The pattern of air flow was consistently as shown in Figure 6 opposite, which shows a vertical cross section through the 6 m deep offices. Cooler air having entered through the lower part of the open windows moves aeross each office at low level and is drawn upwards into thermal plumes above people and heat generating equipment. It then returns to the window end of the office at high level and then passes ‘out through the upper part of the open windows. The pattern of ventilation described above was found to also occur in 12 m deep offices, as the typical example in Figure 7 opposite reveals. This, together with evidence that local ventilation rates and internal temperatures do not vary significantly with distance from the window- end of the office (see Figures 4 and 5), suggests that single-sided ventilation can be applied successfully in offices approaching 12 m deep. Thus, the previously defined limit of 10 m [4] is an acceptable criterion. 5 fi Ee ~~) ---too nek ottice depth e 7 —éem 2 | = -12m ad 7 0 ~ : o 02 04 os 08 1 proportion of distance from window to opposite wall g 32 aes - room nest & 30 : toad (win?) Eo 2 £2 28 = 82 27 a g 2 £ 25 a - a 4 2 4 distance from window (m) Air Distribution in Naturally Venuiated Offices ©BSRIATN 4/09Figure 6 Ventilation airflow pattern - 6m office Figure 7 Ventilation air flow pattern - 12m office Ear (CaSAIA TN 499 ‘Air Distbution in Naturally Ventilated OfficesETE 6 WINDOW TYPE Different window types used in the study are illustrated in Figure 9. The variation of ventilation rate with window type is indicated on Figure 8. The set of data for a window open area of 2.3 m? shows that the windows with openings only at high or low level (sash-low level, and sash-high level) produce lower ventilation rates than the others which either have openings at two levels or, in the case of the horizontal sliding type, have an opening that extends over the full height of the window. However, it was possible to extend the opening of the centre-pivoted and top-hung window types to an area of 3.5 m’, thus producing even higher overall ventilation rates. This illustrates the advantages of window openings at both high and Tow level and of the types that provide large openable areas, = 10 S 2 8 | Figure 8 £ ‘Comparative ie effectiveness of & 4 window types 2 gE 2 = 0 orizontat ‘Air Distribution in Naturaly Ventiated Offices ‘OBSRIATN 4/29Figure 9 Window types Z| Vr LZ Sash/Top open Sash/Bottom open Sash/Top & bottom open Eo Horizontally sliding Centre pivoted Top Hung CBSRIA TN 4199 Air Distibuion in Neturaly Ventated Otfces3.7 VARIATION Figures 10, 11 and 12 show the variations of overall ventilation rate, BETWEEN average room air temperature and average PPD for the different FLOORS window types on each of the three floors included in the computation. ‘The data shown are for one window open area ratio (0.077) and for the lowest internal heat load (14 Wim’). These indicate that the conditions ‘on the upper floors are generally worse than on the ground floor but that satisfactory conditions are much more likely to be achieved at all three levels with the better window types. 10 AirDistribution in Naturally Ventlated Offices easrIA TN 4998 au Figure 10 | = sy Ventilation rates per 25 3 ji floor with window cura = i ‘opening ratio of | & 0.077 androomheat | = 8 load of 14 Wh? & 2 S17 a | aa eae tte oy Bground| =? 2 #5 5f 2s 3 mafirst 5 g 42 ba og Gsecond| &§ § 35 g Figure 11 Average air temperature per floor with window opening ratio of 0.077 and room heat load of 14 Wim? air temperature (deg C) | dissatisfied per floor with window opening ratio of 0.077 and roam heat load of 14 wre = | 3 = Figure 12 = Percentage S < sash - high sash - high ‘and low horizontal ground] mifirst g | Eisecond) & CBSRIATN 4/99 Air Distribution in Naturally Ventlated Offices 11SL SETS 3.8 REQUIRED To determine the extent of window opening required to achieve ‘WINDOW aeceptable internal conditions, the research data were further analysed eee in relation to thermal comfort taking a PPD value of 20% as the upper acceptable limit. Although this may appear to infer that only 80% of the occupants are likely to be comfortable, different personal preferences and the ability of individuals to make adjustments (to their clothing level, for example) means that this criterion will result in a higher level of acceptability that the percentage value might convey On this basis Figures 13 and 14 have been derived for the best window types. These Figures show that for the two window types an opening area ratio of 0.08, as prescribed in Approved Document F of the Building Regulations [2], is adequate for internal heat loads up to about 10 W/m’, Beyond that heat load, additional openable area is required to maintain comfortable conditions. 12 AirDistbution in Naturally Ventlated Offices ©BSAIATN 4199Figure 13 Minimum open area ratio required for centre- pivoted windows igure 14 Minimum open area ratio required for top-hung windows window open area ratio window open area ratio 0.12 o4 0.02 0.12 0.1 0.08 0.06 + 0.04 0.02 ground floor| ‘st floor =—second floor —— 10 15 0-20.25 30 heat load (W/m?) A=L |__| ground floor] -=first floor — second floor| ———— 10 15 20 25 30 heat load (W/m?) CBSRIA TN 4199 Air Distibution in Naturally Ventiated Offices 13Sa ETT] 4 THE STUDY ‘The core of this study was an examination of the thermally-driven natural ventilation in a three-storey office block having 6 m deep offices ventilated only through one external facade. A windless, summer day was assumed with an outdoor air temperature of 24°C. ‘The potential for the achievement of acceptable indoor conditions was assessed for a range of window types and open areas. A section and plan view of the offices is shown in Figure 15, The physical and thermal data is summarised in Table | A similar, though less detailed, study was conducted with 12 m deep offices to assess the performance of single-sided natural ventilation under more extreme circumstances, The analysis was conducted using a commercially available Cl (computational fluid dynamics) code "FLOVENT”. The application of FLOVENT to this type of study was validated by comparing the results ‘obtained in a full-scale physical model in the BSRIA laboratory with those predicted for the same conditions by the software. Having confirmed the viability of using CFD for this application, the PLOVENT software was used to predict the overall and local rates of ventilation as well as room air temperatures and velocities. From this data, analyses of thermal comfort were made on the basis of the ISO Standard [5] model for predicting the percentage of occupants likely to dissatisfied Full details of the research is presented in a BSRIA report "Air distribution in naturally ventilated buildings” [1]. It contains the results of nearly 100 simulations. Table 1 Physical and Thermal Characteristics of the offices Building ‘storeys Each office 6 m deep by 5 m wide by 2.8 high Glazing area ‘4m wide from 1.1 m to 2.98 mhigh = 5.12 m Window types: ‘Sash with bottom openings (maximum open area 2. m") (see Figure ) ‘Sash with top openings (maximum open area 2.3 m’) Sash with top and bottom openings (maximum open area 2.3 m°) Horizontal siding (maximum open area 2.3 m') Centre-pivoted (maximum open area 3.5m) ‘Top hung (both upper and lower panes) (maximum open area 3.5 m*) Direct Solar radiation | 0 (external shading assumed) Uvalue extemal wall | 0.5 Wim? K windows 3.3 Wim? K intemal walis___| No heat transier External conditions Internal heat gains Airtemperature | 24°0 Three occupants. | 300 W maximum Ground surface Three computers | 360 W maximum temperature 208°C Lights 360 W maximum Wind speed o Total values used | 420 W (14 Wim) 660 W (22 Wim’) 800 W (30 Wim?) 14 AirDistouion in Natwaly Ventiated Otfces GASAIA TN 4:99Figure 45 Elevation and plan of simulated office eaSRIA TN 4/99) ‘Air Distibution in Naturaly Ventilated Offices. 15,16 SS TS) 5 REFERENCES i JACKMAN P.J., Air distribution in naturally ventilated buildings. BSRIA Report 77880/2, BSRIA Bracknell, September 1998. 2. Approved Document Fl Means of ventilation. 1995 Edition (The Building Regulations 1991) HMSO, 1994. 3 BS5925 Ventilation principles and designing for natural ventilation. British Standards Institution, London, 1991 4 WALKER R.R., WHITE M.K., Single-sided natural ventilation - how deep an office? Building Serv. Eng. Res. Tech., Vol. 13, No. 4 pp 231-236, 1992. ISO 7730, Moderate thermal environments - determination of the PMV and PPD indices and specification of conditions for thermal comfort. International Standards Organisation, 1984. Air Distribution in Naturally Ventilated Offices cBSAIATN 4199BSRIA is the UK's leading centre for building services research. We offer independent and authoritative research, information, testing and consultancy and market intelligence. ‘Among our clients are consulting engineers, contractors, manufacturers, building operators, government bodies and utlities. We work closely with these clients, taking full account of individual priorities and needs, and maintaining individual confidentiality at all times. Our specialist skills, knowledge and facilities will complement your expertise at every stage of the building process. Founded over 40 years ago, BSRIA now has a staff of 120 and operates from two well equipped laboratories and office premises in Bracknell and Crowthorne. As a member-based organisation we also provide a focus for cooperative research, offering a partnership between industry and government. The Association offers: ¢ Awide range of member services ¢ Accollaborative research programme focused on industry needs Confidential contract research facilties, tailored to your requirements ¢ Independent equipment and systems testing, investigation and consultancy ‘© Market research consultancy, publications and information for the UK and overseas markets ‘# Instrument hire and calibration services. ‘Our technology centres offer research and consultancy on: ‘+ Domestic, commercial and industrial heating Mechanical and natural ventilation, air infitration Air conditioning and refrigeration Plumbing and drainage systems Water quality Electrical services Lighting Building management systems and controls Fire and security systems wee eee res Environmental issues. For further details contact Marketing Services at BSRIA The Building Services Research and Information Association Old Bracknell Lane West, Bracknell, Berkshire RG12 7AH UK Tel: +44 (0) 1344 426511 Fax: +44 (0) 1844 487575, e-mail: bsria@bsria.co.uk web: www.bsria.co.uk