Sdar journal 2013 web copy by Pat Lehane

Journal

Sustainable Design & Applied Research in Engineering of the Built Environment November 2013 Issue 3

The Journal of Applied Research in Innovative Engineering of the Built Environment

Building Servicesnews

Successful businesses are always open to

controlling energy costs. At Electric Ireland, we understand that reducing costs is essential to the success of any business. We also know that, when it comes to payment plans, one size doesn’t fit all. That’s why we offer you very competitive fixed or variable rates that take account of what your individual company needs. Electric Ireland have a long history of investing in the Irish electricity network and are continually looking for ways to help our customers control their costs. Talk to our dedicated business team about how we can help you reduce your costs and use energy more efficiently.

Call the business team today on

1850 30 50 70 electricireland.ie/business

Ireland’s energy 156319_EI_SME_Florists_BizEyeMag_ROI_FA_BD.indd 1

11/10/2013 15:51

Contents

Editor’s foreword

A reader’s guide to this issue

Rethinking daylighting and compliance John Mardaljevic

LEDs are the panacea – and other fairytales James Thomas Duff and Peter Whitty

Wind resource in the urban environment Jonathan Blackledge, Derek Kearney and Eamonn Murphy

Introduction

Welcome to the third edition of the SDAR Journal which the Chartered Institution of Buildings Services Engineers (CIBSE) in Ireland is delighted to copublish with DIT. This is an excellent source of information for engineers, researchers, designers and all involved in the built environment and is a fine example of industry and third-level colleges working together to provide research in the areas of sustainability and low energy technology. The challenges we face today of climate change, rising energy costs and general economic difficulties require innovative solutions that only can be delivered by detailed investigation of new ideas and new technologies. In Ireland we are fortunate to have highly-educated and capable engineers and scientists that can deliver on the demand for more research. Our collaborative approach with DIT means we help to harness that.

The Small Wind Energy Estimation Tool (SWEET) – a practical application for a complicated resource

The various events that CIBSE Ireland organises in conjunction with the DIT are all potential sources for future research papers, while CPD events and the masterclass at the SEAI Energy Show are also ideal platforms to showcase applied research.

Keith Sunderland, Thomas Woolmington, Gerald Mills, Jonathan Blackledge and Michael Conlon

I would encourage third and fourth level research students, academic staff, and also the industry as a whole, to participate in these events by submitting ideas, abstracts and papers, including case study information, that we can all learn from.

Performance of a demand-controlled mechanical extract ventilation system for dwellings

DIT and CIBSE Ireland have a strong record of supporting potential industry authors and are committed to continuing this, especially with research papers that can be included in future editions of the SDAR Journal.

I. Pollet, J. Laverge, A. Vens, F. Losfeld, M. Reeves and A. Janssens

Sean Dowd Chairman, CIBSE Ireland

The SDAR Journal is a sustainable design and applied research publication written by engineers and researchers for professionals in the built environment Editor: Dr Kevin Kelly, DIT & CIBSE Contact: kevin.kelly@dit.ie Support Editorial Team: Thomas Shannon, Yvonne Desmond, Pat Lehane Reviewing panel: Professor Gerald Farrell, Professor Eugene Coyle, Dr John McGrory, Sean Dowd, Michael McNerney, Derek Mowlds, Brian Geraghty, Kevin Gaughan, Dr Martin Barrett, Dr Marek Rebow, Dr Michael Conlon, Dr Mike Murphy, Brendan Keely, Michael Farrell and Professor Tim Dwyer Upload papers and access articles online: http://arrow.dit.ie/sdar/ Published by: CIBSE Ireland and the School of Electrical & Electronic Engineering, DIT Produced by: Pressline Ltd, Carraig Court, George’s Avenue, Blackrock, Co Dublin. Tel: 01 - 288 5001/2/3 Fax: 01 - 288 6966 email: pat@pressline.ie Printed by: Swift Print Solutions (SPS)

DIT is delighted to be co-publisher of the third edition of this journal and the ongoing collaboration between DIT and CIBSE/SLL in this publication is important to us as an academic community. We are also proud to have supported CIBSE/SLL in organising the 2013 International Lighting Conference and we look forward to welcoming the 2014 CIBSE Technical Symposium to DIT in April 2014. As Head of School of Electrical and Electronic Engineering in DIT I am very aware how important it is for us to continue to build strong industry relationships and to offer a platform to share significant research contributions with the world of engineering and the built environment. After a recent merger our school is now the largest school of its kind in the country with over 1200 students. We focus on applied research which is recognised for its impact and quality, which in many cases is on a par with that of the very best groups internationally. We have a strong emphasis on research in energy management and renewable technologies, and the influence of researchers in our school in this area is evident in the research carried out and in the support for this journal provided by the school. I would encourage engineers in industry to collaborate with our school in publishing more of their work through this journal, and in entering the SDAR Awards (see page 52) and Irish Lighter Awards (see page 51).

Professor Gerald Farrell

Additional copies can be purchased for €50

Head of School of Electrical & Electronic Engineering DIT

CIBSE Sustainable Awards 2013

Editor’s foreword

A Reader’s Guide

Dear reader, welcome

We attempt in this issue to provide a series of articles

to the third of what has become an annual journal published by CIBSE Ireland and the School of Electrical & Electronic Engineering in DIT. This is a successful collaboration between a community of professional building services engineers with academics and researchers in the largest higher education institute in Ireland. We will maintain a once-a-year publishing frequency while the number of entries and level of papers remains at its present rate, but we may increase the frequency of publication if the number of goodquality entries increases. Up to now, papers have been mainly by Irish authors about Irish projects. Initially we wanted to hear about the good, the bad and the ugly of the application of low energy projects in the built environment in Ireland. The intention was to encourage applied research and postoccupancy evaluation in low energy engineering of construction projects, publish the results to disseminate insightful findings to the industry, and so help improve practice on the ground in Ireland. To a large extent that is still our main objective but we also want to encourage more of our many international readers to submit papers. This current edition includes our first international papers. One is a joint paper from Belgium and the UK, which featured in the CIBSE Symposium in Liverpool in 2013. This might whet the appetite for the 2014 CIBSE/ASHRAE International Symposium to be held in DIT, Kevin Street Campus, Dublin, on 3/4 April 2014. The other international paper is also from the UK, from a world-renowned expert on daylighting of buildings. If you have post-occupancy evaluation data, interesting feedback on low energy installations, or new ideas you have tried and evaluated, then please submit your proposals online to http://arrow.dit.ie/sdar/. All our previously-published papers are there alongside papers from the 2013 International Lighting Conference from Dublin. We would be delighted to receive your abstracts or ideas and can offer assistance and support in writing up papers. Working engineers in industry have access to data and are often data rich and time poor, while researchers in academia are data poor but sometimes have some time for research. We in DIT will help you interrogate your data in order to publish a scholarly paper. If you would like to explore this then please contact me at kevin.kelly@dit.ie

Kevin T. Kelly C Eng FCIBSE FSLL FIEI President SLL; Head of School of Multidisciplinary Technologies, College of Engineering & The Built Environment, Dublin Institute of Technology.

combining some of the main pillars of low energy building design. Papers on daylighting, mechanical ventilation, two on micro wind generation and finally one on LED lighting come together in an eclectic mix that we think should provide interest across the whole building services sector. The first paper is on daylighting and was presented at the international lighting conference in Croke Park, Dublin in April 2013. Professor John Mardaljevic from Loughborough University questions our uncritical use of assumptions, guidelines and rules-of-thumb that have applied for many years in the calculation of daylight factor in buildings. He examines current practice with respect to daylight calculation and suggests ways he thinks would improve this practice which are well argued and would not prove too onerous on the community if adopted. This paper is essential reading for any lighting designer, architect or electrical engineer, but is also very readable for those of other disciplines who would like to see how simply daylight levels inside a building can be calculated. Duff and Whitty address the assumption that LEDs are the panacea for low energy lighting solutions. They explain the growth of LEDs, examine how LEDS can be evaluated, and explain how lighting engineers might protect their client’s, and their own reputations, from the risks associated with some LED products. They bring forward a proposal for a standard set of questions to which they believe any reputable manufacturer should be able to respond and, more importantly and uniquely, a set of what they suggest are acceptable responses. This would facilitate lighting designers and those specifying lighting products to make fair comparison between different manufacturers. With the demand for renewable technologies for new buildings over 1000m2, we publish two papers on wind. The first by Blackledge, Kearney and Murphy reports on

1-3 Intro pages 2013:Layout 1

17/10/2013

09:05

Page 3

A Readerâ&#x20AC;&#x2122;s Gude

a project monitoring wind in the DIT Kevin Street Campus. While it is not unexpected that wind will vary, accurately measuring and recording the wind resource is very challenging. Readers will get an insight into turbulence, boundary layers and the difficulties encountered. This led the researchers to the development of a prototype anemometer that is expected to address the need for digitally mapping real-time threedimensional data on wind. The second paper on wind is also from PhD research in DIT in collaboration with University College Dublin. This paper is entirely separate from the first but interestingly complements it. It uses detailed wind observations to model turbulence characteristics. The results of this are incorporated into a practical Excel tool so that engineers can gain an intuitive appreciation of the benefits and limitations of the wind resource for their own projects and thus enable them make informed decisions and install wind turbines when it is appropriate to do so. The final paper is from the mechanical engineering side of our discipline but comes from abroad. Pollet et al investigate demand-controlled mechanical extract ventilation based on natural supply in habitable rooms and mechanical extraction in wet rooms. Indoor air quality and energy consumption are determined. The findings are that DCV systems can significantly improve air quality while reducing energy consumption when compared with mechanical ventilation systems. Primary heating energy is reduced and fan consumption is less. Total operational energy is similar to mechanical ventilation heat recovery but is cheaper and requires less maintenance.

The SDAR Awards is a joint initiative between CIBSE Ireland and DIT, sponsored by John Sisk & Son and supported by Building Services News. The awards are unique in that they are intended to disseminate knowledge, encourage research in sustainable engineering of the built environment and raise the quality of innovation and evaluation in such projects. Entries are required to critically evaluate real-life data, and examine both successes and challenges within leading-edge projects throughout Ireland or further afield.

arrow.dit.ie/sdar/

Initial submissions for the SDAR Awards 2014 â&#x20AC;&#x201C; comprising short abstracts of between 100 and 200 words â&#x20AC;&#x201C; should be sent directly to Michael McDonald at michael.mcdonald@dit.ie and/or Kevin Kelly at kevin.kelly@dit.ie. They must arrive by close of business on Monday, 16 December 2013. See page 52 for more details.

For further information contact: michael.mcdonald@dit.ie or kevin.kelly@dit.ie

Email: contact@cibseireland.org Web: www.cibseireland.org

We’re always on the lookout for you … CIBSE is the professional body that exists to “support the science, art and practice of building services engineering, by providing our members and the public with first class information and education services and promoting the spirit of fellowship which guides our work.” CIBSE promotes the career of building services engineers by accrediting courses of study in further and higher education. It also approves work-based training programmes and provides routes to full professional registration and membership, including Chartered Engineer, Incorporated Engineer and Engineering Technician. Once you are qualified, CIBSE offers you a range of services, all focussed on maintaining and enhancing professional excellence throughout your career. CIBSE members in Ireland are represented by an active Regional Committee which is involved in organising CPD events, technical evenings, training courses, social events and an annual conference. The committee welcomes new members, suggestions, and collaborations with other institutions in the built environment.

Providing best practice advice, information and education services The Chartered Institution of Building Services Engineers in Ireland

Rethinking daylighting and compliance

John Mardaljevic School of Civil and Building Engineering, Loughborough University, Loughborough, Leicestershire, LE11 3TU, UK

j.mardaljevic@lboro.ac.uk

Building Servicesnews

CIBSE Sustainable Awards 2013

Abstract Daylight in buildings is the natural illumination

1. Background

experienced by the occupants of any man-made

Towards the end of the 1990s the daylighting of buildings began to achieve greater attention than had previously been the case. There were a number of reasons for this, but the two most important “drivers” were:

construction with openings to the outside. Our attempts to formulate some measure of daylight provision in buildings can be traced back over a century, and the daylight factor as we know it today is over 50 years old. Still the most common measure found in guidelines and recommendations worldwide, the daylight factor is used routinely and, it is fair to say, often rather uncritically. The consideration of daylight in buildings has received a new impetus from the accumulation of evidence on the wider benefits of daylight exposure. But it is continuing to prove difficult to advance beyond daylight factors towards a more realistic quantification of daylighting performance that would allow us to accommodate these new considerations in an evaluative schema. This paper examines the basis of current practice with respect to daylight evaluation, and suggests a few ways in which it can be improved with relativelymodest additional effort. The paper also critiques some of the recent attempts to advance daylight evaluation by incremental means using so-called “clear sky options”.

Key Words: Daylight standards, daylight modelling, CIE overcast sky, daylight factor.

1. the widespread belief that the potential to save energy through effective daylighting was greatly under-exploited; and, 2. the emergence of data suggesting that daylight exposure has many positive productivity, health and well-being outcomes for building occupants. The first originated with the widely-accepted need to reduce carbon emissions from buildings in order to minimise the anticipated magnitude of anthropogenic climate change. This in turn led to the formulation of guides and recommendations to encourage the design and construction of “low energy” buildings and also for the retrofit of existing buildings. All these guides contain recommendations on daylighting, invariably founded on the daylight factor or an equally-simplistic schema such as glazing factors [1]. The second driver was the gradual accumulation of data from disparate sources on the non-visual effects of daylight exposure. These effects are believed to be wide-ranging and include productivity and health/well-being, e.g., academic achievement, retail sales, recovery in hospitals, entrainment of circadian rhythm, etc. The mechanisms for these effects are not yet fully understood, and it is not yet known what the preferred exposure levels should be, nor if existing guidelines would be effective for these quantities [2]. Nonetheless, given the still relatively-low cost of electric lighting – and the potential for it to be further reduced with solid-state lighting – there is evidence to suggest that the cost-benefit from increased productivity due to good daylighting could be far greater than the cost-saving from reduced-energy expenditure [3]. Thus the second of these drivers has been promoted on both economic and environmental quality grounds. Almost concurrent with the emergence of the two key drivers noted above were a major advance in the way daylight in buildings could be modelled, and the development of numerous new glazing systems and materials to better exploit daylighting in buildings. These developments are expected to lead to significant changes in the way that daylight in buildings is both evaluated, i.e., using climate-based daylight modelling (CBDM), and exploited, e.g., by new glazing systems and materials. It should be noted that the performance of these new glazing systems often depends on their ability to shade and/or redirect sunlight. Thus they can only be reliably assessed using CBDM – the standard approaches (e.g., daylight factor) are unsuited to the task. Notwithstanding that it is over a decade since CBDM was first demonstrated and its effectiveness subsequently proven on a range

Rethinking daylighting and compliance

of “real world” projects, daylight criteria in most guidelines and recommendations are still founded on the daylight factor. More recently there have been attempts to advance the DF method incrementally using so-called “clear sky” evaluations, though these appear unsatisfactory for the following reasons. There are a number of reasons why it has proven difficult to advance towards metrics founded on climate-based daylight modelling. Perhaps part of the difficulty in effecting this “journey” is that we are not entirely certain regarding the point of departure – what exactly is the basis and rationale for the ubiquitous daylight factor method? This paper dissects both the basis of the method (i.e., relative values predicted using the CIE standard overcast sky) and how it is often applied to characterise a space, e.g., by giving an average daylight factor value. The three sections that follow the note below are “vignettes” of how I imagine the much-needed deeper discussions on these matters might progress. The logic presented is mainly by way of argument illustrated with a handful of examples. The tests required to definitively confirm or disprove the various hypotheses framed here are beyond the scope of a relatively brief paper. However, I hope that the reader will find the propositions sufficiently intriguing to engender further debate on these matters.

1.1

A note on the origin and formulation of the daylight factor

to be defined since the luminance distribution of the sky will influence the value of the ratio. At the time that the daylight factor was first proposed it was assumed that heavily-overcast skies exhibited only moderate variation in brightness across the sky dome, and so they could be considered to be of constant (i.e., uniform) luminance. Measurements revealed however that a densely-overcast sky exhibits a relative gradation from darker horizon to brighter zenith; this was recorded in 1901. With improved, more sensitive measuring apparatus, it was shown that the zenith luminance is often three times greater than the horizon luminance for some of the most heavily-overcast skies [5]. A new formulation for the luminance pattern of overcast skies was presented by Moon and Spencer in 1942, and it was adopted as a standard by the CIE in 1955. Normalised to the zenith luminance Lz , the luminance distribution of the CIE standard overcast sky has the form: Lθ =

Lz (1 + 2 sin θ) 3

where Lθ is the luminance at an angle θ from the horizon and Lz is the zenith luminance (Figure 1).

Being mean to the average

It appears that the daylight factor, or at least its precursor, was first proposed in 1895 by Alexander Pelham Trotter (1857-1947) [4]. The origins of the daylight factor are actually somewhat hazy since there appears not to have been a seminal paper introducing the approach. The reference to its introduction in 1895 appears to be anecdotal and recalled a number of years later.

It is proposed here that the average should no longer be used as a means to summarise measures such as the daylight factor distribution. The average is typically used to give a “bottom line” number which is intended to be the sole daylight performance indicator for the space. Instead, the median (or a quartile) should be employed whenever a single quantity is required to characterise a space.

The daylight factor was conceived as a means of rating daylighting performance independently of the actually-occurring, instantaneous sky conditions. Hence it was defined as the ratio of the internal horizontal illuminance Ein to the unobstructed (external) horizontal illuminance Eout, usually expressed as a percentage, Figure 1. However, the external conditions still need

The average tells us nothing about the distribution of DF in the space, whereas the median does. The average can be a quite misleading quantity when applied to daylight distributions, especially for spaces illuminated from vertical glazing on one wall where the very high DFs close to the windows can significantly influence the average DF value. Because of this, the average is very

Figure 1: Definition of the daylight factor and the CIE standard overcast sky

CIBSE Sustainable Awards 2013

sensitive to the proximity of the sensor plane to the glazing. The closer the sensor points are to the glazing, the higher the average for the daylight factor distribution in the space. As far as I am aware, it was not until the appearance of the 2011 revision of Lighting Guide 5 (LG5) that a recommendation for a perimeter zone between sensor points and glazing/walls has been given in a UK guideline for simulation. LG5 recommends a 0.5m gap (perimeter zone) – which seems reasonable, though it should be noted that the rationale given in LG5 (i.e. to avoid the low values at the back of the space) is incorrect. The upper section of the plot shown in Figure 2 gives the DF distribution across (half) of a 6m wide by 9m deep side-lit space (2.7m floor to ceiling height). Here the sensor plane covers the entire 6m by 9m internal plan, though the glazing is located on the outer side of a 0.2m reveal (so it is not quite a worst-case regarding close proximity of the sensor plane to the glazing). The average DF for this scenario was 2.8%, however the median value was only 1.1%. With the latter we know that only half the area of the space has achieved a DF of 1.1%, whereas with the average we have no such certainty. More worryingly, the average can in some people’s minds be conflated with the median, thereby giving a completely false impression of the DF distribution for the space. Having a reasonable perimeter (e.g., 0.5m) reduces the size of the false impression given by the average, but it does not eliminate it.

climate for Abu Dhabi reveals that it is almost never overcast in that region of the United Arab Emirates. This, not unexpected observation, suggests that at least in some instances the daylight factor has indeed been applied well outside of its “zone of applicability”, notwithstanding the uncertainty regarding its precise boundaries. The link between (relative) daylight factors (i.e. percentage values) and absolute levels of illuminance (i.e. lux) has always been tenuous. DFs are of course derived from absolute values, but the latter are often ignored thereafter. Design guides often give recommendations in terms of daylight factors, but then also suggest that daylight should provide illuminances of say 300 lux or more for much of the year. Building Bulletin 90 (Lighting Design for Schools) does describe how to relate DFs to estimates of achieved absolute illuminance [7], but these “conversions” are rarely carried out.

CIE overcast sky 10%

Uniform sky

The implication being that, if we provide a certain measure of daylight for the “worst case”, then surely it can only be better than that for the rest of the time. However, while such notions are suggestive, the rationale for the daylight factor has rarely, if ever, been rigorously expounded. For example, what exactly is meant by “worst case”? Is it that the absolute values provided by the sky (i.e., the diffuse horizontal illuminance) is (are?) “worst case”, or is it perhaps that the luminance distribution on the sky vault is a “worst case”? Or maybe a combination of the two? Moreover, if the daylight factor is suitable for “northern Europe”, what is the extent of its zone of applicability? The daylight evaluation in the first edition of the Estidama Pearls Design System for Abu Dhabi was founded on daylight factors, i.e., the CIE standard overcast sky [6]. A quick examination of the standard

CIE overcast sky

A gloomy view of the CIE overcast sky

At first glance, the CIE overcast sky seems a reasonable basis for the evaluation of daylight in buildings. This “feeling” is perhaps formed, or at least bolstered, from seeing phrases such as these commonly found in documents pertaining to daylight evaluation: “the overcast sky represents worst case conditions”; “the daylight factor is defined as the worst case”; and, “the daylight factor was invented in northern Europe where the fully overcast sky is common”.

0.5%

Sensor plane at desk height 10%

In contrast, the median value is largely insensitive to the size of the perimeter, and so it is not only a more informative measure, it is also more robust since it is largely unaffected by any “game playing” with respect to the placement and size of the perimeter.

Sensor plane at ﬂoor level

Uniform sky 9m

Figure 2: Daylight factor distribution for CIE standard overcast and uniform skies

In Australia and New Zealand a uniform sky is used for what are in effect “daylight factor” calculations, though the sensor plane is set at floor level rather than at desk height, introducing another dissimilarity when comparing methods. The differences in predicted distributions between “classic” daylight factor (i.e., overcast sky and sensor plane at desk height) and the option recommended in Australia/NZ (i.e., uniform sky with sensor plane on the floor) is revealed by comparing the plots shown in Figure 2. The metrics derived from each of the four distributions are given in Table 1.

Rethinking daylighting and compliance

Table 1: Metrics derived from the distributions shown in Figure 2 Sky type (sensor height)

Average [%] Median [%]

Max [%] Min [%]

CIE overcast (desk) Uniform (desk)

2.8 3.4

1.1 1.6

15.3 15.9

0.38 0.60

CIE overcast (floor) Uniform (floor)

2.5 2.9

1.6 2.0

8.1 8.3

0.47 0.71

As we might expect for a side-lit space, a uniform sky produces higher ratios (i.e., DFs) than a CIE overcast because, with the latter, the sky vault luminance is “concentrated” around the zenith – the average, median and minimum DFs are markedly higher for the uniform sky, Table 1. Placing the sensor plane at floor level results in a vastly-different DF pattern compared to when at desk height – irrespective of the sky type, Figure 2. Because of the sill, the window from the perspective of a sensor plane on the floor appears more like a clerestory window, i.e., the peak in both DF distributions is displaced away from window rather than closest to it. Note that, if the glazing in the space was floorto-ceiling, then the DF distributions at floor level would appear similar to those in the upper plot, but with higher absolute values since the sensors now “see” a greater expanse of sky. Furthermore, the values would be highly misleading because the sensors now include the contribution of light that enters the space below desk height, i.e., heading directly for the (typically) lowreflectance floor where most of the light will be immediately absorbed. I have not been able to locate any documents that describe the rationale for placing the sensor plane at floor level. While it is hard to see any benefit in having the sensor plane at floor level, might there nevertheless be a case for basing estimates of internal illuminance availability on the uniform rather than the standard overcast sky? Consider the cumulative diffuse availability curve shown in Figure 3. The following estimates can be derived from the curve: a 2% DF gives ∼100 lux for 85% of the year, whereas the same DF gives an illuminance of ∼300 lux for about 55% of the year. Occupants will, of course, invariably prefer daylight illuminances around the 300 lux mark compared to those around 100 lux. Some of the skies around the 5,000 lux (diffuse horizontal illuminance) mark are likely to conform to varying degrees to the standard overcast pattern. However, the 15,000 lux skies needed to produce 300 lux internally (for a DF of 2%) are much more likely to have luminance distributions that diverge significantly from the standard overcast pattern. That will be even more the case for the remaining 45% of the skies in the distribution that have higher diffuse illuminances. In other words, when the DFs are predicted using the standard overcast sky, the basis for the estimate of the occurrence of internal illuminances is strictly self-consistent only for those skies in the annual climate file that conform to the CIE standard overcast pattern. But what is the proportion of annual occurring skies that are a good match for the CIE standard overcast sky? That is not an easy

question to answer. It is possible to determine the annual occurrence of essentially overcast skies in standard climate datasets using, say, the Perez clearness index. However, those will include the whole gamut of overcast skies, many of which it seems do not exactly match the standard pattern: Enarun and Littlefair suggest that “… if a general cloudy sky is all that is required, the CIE may not be the best option” [8]. In the same paper they suggest that the “quasi-overcast sky” may serve better as a “general cloudy sky”. The quasi-overcast sky has a more gradual gradation between horizon and zenith compared to the CIE standard overcast. But, it also includes a small component which varies with angle from the sun. Thus, it could not replace the use of the CIE standard overcast in a daylight factor evaluation because the sun position is now a factor in the evaluation. To recap: The CIE standard overcast sky is in fact – to quote Enarun and Littlefair – an “extreme” case of overcast sky. Thus, skies that conform to the CIE standard overcast sky pattern are likely to be rarer than is generally imagined, and in any case produce internal illuminances at, or below, the lower end of what is generally preferred by occupants. A sky luminance distribution with a smaller ratio between horizon and zenith is believed to be a better fit to the more typical gamut of overcast skies, i.e., the brighter overcast skies that deliver more useful levels of natural light for occupants. Given that the “quasi-overcast” cannot replace the CIE standard overcast in a DF-based evaluation, perhaps the uniform sky is in fact the “best” simple sky condition on which to base estimates of daylight provision using diffuse illuminance curves. In fact, the uniform sky is probably a closer fit to an average of the “quasiovercast” (for varying sun positions) than the CIE standard overcast pattern. Furthermore, it is perhaps not unreasonable to describe the CIE standard overcast sky pattern as one that exhibits bias when used to estimate the occurrence of internal illuminance from DFs. This is because the luminance pattern – maximum at the zenith – deviates from the gamut of typically-occurring overcast patterns in a consistent manner. The effect on the prediction of ratios at the sensor plane (i.e. DFs) is evident in Figure 2 and Table 1. Compared to the uniform sky, the maximum DFs for the standard overcast sky are more tightly packed closer to the window where the sensors have the best “view” of high altitude sky close to the zenith. The case for suggesting that a uniform sky might actually be a sounder basis for daylight design than the CIE standard overcast can be reasoned, as demonstrated above. But it is not at all clear at this stage how it might be tested. In large part this is because we do not yet have a robust notion regarding an agreed-upon datum against which we can discriminate outcomes. A somewhat idealised datum is of course some measure that, if achieved, ensures “good daylighting”. One proposed measure made by the IES Daylight Metrics Committee is the annual occurrence of 300 lux across the workplane.

% off working year diffuse illum illuminance minance exceeded

CIBSE Sustainable Awards 2013

100

Limited number standard overcast

Skies increasingly less likely to be overcast

0 0

10 0

30 40 illumin nance (klux) Diffuse illuminance

2% 2% DF DF gives gives ~300lux ~300lux for for ~55% ~55% time time 2% 2 % DF DF g gives ives ~100lux ~100lux for for ~85% ~85% time time Figure 3: Cumulative curve showing diffuse illuminance availability

New approaches: Do ‘halfway’ methods work?

The “clear sky option” appears to have been introduced in LEED version 2.2 as a means to overcome the limitations of the climate/orientation insensitive glazing factor and daylight factor methods. To achieve credit 8.1, the requirement can be: Demonstrate, through computer simulation, that a minimum daylight illumination level of 25 footcandles has been achieved in a minimum of 75% of all regularly occupied areas. Modelling must demonstrate 25 horizontal footcandles under clear sky conditions, at noon, on the equinox, at 30 inches above the floor. While this may appear reasonable at first, the LEED v2.2 documentation gives no supplementary data for the evaluation. This omission all but renders the evaluation meaningless since there is no statement regarding the diffuse horizontal illuminance that the sky should be normalised against. The user, it seems, is to trust the default value that is provided by the sky generator program. The default value is an extremely coarse approximation with some latitude dependance (and of course time of day/year), but no basis whatsoever in local, prevailing climatic conditions. Many users are unaware that the key input parameter for their simulation is of dubious provenance and has been automatically

selected on their behalf. It gets worse, as there is no mention of what the sun luminance (usually derived from direct normal illuminance) should be. This too is surprising, since the sun contribution will greatly add to the illuminances resulting from the diffuse sky (which will depend on the unspecified diffuse horizontal illuminance anyway). Given the relatively modest target illuminance (around 250 lux) it seems likely that the evaluation is meant to be carried out using a clear sky distribution without a sun. This, of course, is a physical impossibility in reality. Anecdotal evidence has confirmed that users of LEED have indeed “demonstrated compliance” with the recommendations and obtained Daylight Credit 8.1 by using a physically-impossible luminous environment (i.e., clear sky without sun) that is normalised to an unknown diffuse horizontal illuminance. ASHRAE Standard 189.1 (2009) has a similar clear sky option to LEED. As with LEED, there is no mention of normalising the sky to a specified diffuse horizontal illuminance, so the same shortcomings (outlined above) apply. As with the LEED Clear Sky option, the ASHRAE draft guidelines suggest (but do not clearly state) that the clear sky modelling is to be done without a sun – which is, as noted above, a physically impossible illumination condition in nature.

Rethinking daylighting and compliance

The ASHRAE draft states that either the CIE Overcast or the CIE Clear sky model may be used. This offers intriguing possibilities to the artful compliance chaser, since the outcome it turns out depends to a large degree on what default values “drop out” of the sky generator program. Since many practitioners use the Radiance lighting system, either in its raw (UNIX) form or in one of the many bundled packages, it’s instructive to see how different the outcomes can be depending on the choice of sky. The Radiance (UNIX) command “gensky 3 20 12 –c” generates a description of the brightness distribution of the CIE standard overcast sky for noon, 20 March (i.e., month 3). A similar command generates the description for the CIE clear sky pattern. The guidance gives no recommendation regarding normalisation of the skies to a known diffuse horizontal illuminance (Edh). So, the diffuse horizontal illuminance of the resulting sky depends entirely on how the sky model generator program gensky was devised to produce skies of either type, i.e., its default behaviour. The diffuse horizontal illuminance for the two sky types turns out to be very different – almost by a factor of two, Table 2. It may even seem counterintuitive that, without any user intervention by way of supplying normalisation values, the diffuse horizontal illuminance for the overcast sky should be nearly twice that of the clear sky.

Table 2: Diffuse horizontal illuminance (Edh) for standard overcast and standard clear skies generated without normalisation CIE sky type

Radiance command

Edh [lux]

Standard overcast

gensky 3 20 12 –c gensky 3 20 12 –c

14,679

Standard clear

8,454

However, the reason is quite straightforward. The sky model generator program does not have any knowledge of local meteorological conditions. What it does know are: latitude/ longitude i.e., location (the default of Berkeley, USA is used in the example), time of day/year and therefore sun position, and also the incident extraterrestrial solar radiation. This last part is apportioned between sky and sun (if present) depending on the selected sky type. For an overcast sky the incident extraterrestrial solar radiation is reprocessed into diffuse sky radiation (using default values for turbidity etc). But, for a clear sky distribution, the extraterrestrial solar radiation is apportioned between the sky radiation and the (now expected but not included) sun. Thus, the diffuse horizontal illuminance for the clear sky is lower (typically just over half using the gensky defaults) than the diffuse horizontal illuminance for the overcast sky. Notwithstanding the differences in the sky luminance patterns, the designer “chasing” the attainment of an absolute level of interior illuminance would be advised to opt for the overcast sky because of its much higher diffuse horizontal illuminance. In recognition of what must be viewed as a less-than-ideal state of affairs regarding the lack of normalisation in the “clear sky option” of Version 2.2, the 2nd Public Comment Draft on LEED (July 2011) contains the following:

Demonstrate through computer simulations that the applicable spaces achieve illuminance levels between 300 lux and 3000 lux for both of the following sky conditions: – 9:00 am equinox on a clear-sky day (solar time) – 3:00 pm equinox on a clear-sky day Illuminance intensity for sun (direct component) and sky (diffuse component) for clear sky and overcast conditions for those time periods shall be derived from the local weather data, or TMY weather tapes for the nearest city, first by selecting the two days within 15 days of September 21st and March 21st that represent the clearest sky and most overcast sky condition, and then averaging the hourly value for the appropriate spring and fall hour. While this revision might, at first, be seen as “heading in the right direction”, it too has potential problems and confounding issues. The patterns of hourly values in the illuminance datasets are unique and, because of the random nature of weather, they will never be repeated in precisely that way, Figure 4 (see next page). Climate datasets are, however, representative of the prevailing conditions measured at the site, and they do exhibit much of the full range in variation that typically occurs, i.e., they provide definitive yardstick quantities for modelling purposes – provided that the entire year is used in the evaluation. The solid lines on the plots in Figure 4 mark the times of the equinoxes, and the dashed lines mark the date 15 days either side of each equinox. As is evident from the pattern, while it might be likely that a sunny (i.e., clear sky) period occurs within 15 days either side of the equinox, it is by no means certain because of the random nature of weather. Also, how “clear” is clear? That is not specified. Thus, it is highly problematic to attempt to extract and define supposed “representative” illuminance data from climate files. Furthermore, “averaging” of any climatic illuminance data is risky since the user must ensure that the conditions to be averaged are indeed similar. Based on the attempts made thus far, it does not seem possible to advance the DF approach by incremental means, i.e., evaluations based on “clear sky options”, “snap-shots” or “salami-slicing” of climate data. Efforts in this direction have resulted in methods that are one or more of the following: confusing, inconsistent, prone to the vagaries of patterns in climate data, and/or without a proven rationale. There seems to be no half-way house between a DF-based evaluation (e.g., in conjunction with cumulative diffuse illuminance curves) and a fullblown annual evaluation using climate-based daylight modelling.

Discussion

Notwithstanding the more than occasional tone of a jeremiad, this article is in fact intended to accentuate the positive – we do have the means now with climate-based modelling to greatly advance the basis of daylight evaluation for buildings. However, CBDM tools are still largely the preserve of lighting simulation experts/researchers. For CBDM to become mainstream, the software to do it needs to be taken up and supported by one or more major software houses.

CIBSE Sustainable Awards 2013

Diffuse Horizontal Illuminance 24 20

Hour

16 12 8

lux

70000 60000

0 1

6 7 Month

50000 40000 30000

Direct Normal Illuminance 24 20

20000 10000 0

Hour

16 12

Figure 4: Illuminance data from the standardised climate file for London

Herein lies a classic “chicken and egg” conundrum. On one hand, those who draft guidelines are reluctant to recommend metrics founded on CBDM because tools to predict the metrics are generally not available, at least not as supported software. On the other hand, the software vendors are understandably loathe to dedicate the resources to develop and maintain CBDM tools because – inasmuch as climate-based metrics are not in the guidelines – there will be no real market for these new tools. This presents something of an impasse to all those who strive to advance daylighting standards beyond the current guidelines. A suggested way around this impasse follows. In order to obtain “buy-in” from all stakeholders (e.g., standards bodies, designers, end-users, tool developers, etc) it is important that first they recognise the benefit of the changes proposed. These benefits should include the following: • a more robust approach to evaluating daylight in buildings using existing tools with only modest enhancements; • a methodology that allows for later progression to more realitybased evaluations; • and, a transition roadmap with clear market horizons to ensure that software vendors invest the necessary resources to develop the next generation of modelling tools (i.e., CBDM for ‘endusers’).

To this end, it is proposed that current standards based on daylight factors should be upgraded as soon as possible to evaluations founded on the annual occurrence of an absolute value for illuminance (e.g., 300 lux) estimated from the cumulative availability of diffuse illuminance as determined from standardised climate files. For example, a daylighting “target” could be that half of the sensor points in a side-lit space should achieve 300 lux for half of the time when the sun is above the horizon. This is an application of an established but largely-neglected approach [9]. Such an upgrade requires only a modest extension to existing DF software and, importantly, it provides some “connectivity” between the daylight availability and the prevailing climate. Note also that there may be good cause to use a uniform rather than a standard overcast sky for this evaluation. Of course, this is not a full-blown climate-based solution since direct/indirect sun is not accounted for. However, unlike the “halfway” measures described in this article, the cumulative illuminance approach has a defensible rationale. Furthermore, by shifting the analysis to measures of absolute illuminance, it prepares the ground for a relatively smooth transition to eventual, full-blown CBDM evaluations. One could envisage, say, a three year “overlap” period in standards/guidelines during which either the

Rethinking daylighting and compliance

cumulative illuminance approach or CBDM could be used to demonstrate compliance. Then, at the end of that period, only evaluations founded on CBDM would be permitted. Such a provision would encourage software houses to invest the time and resources to develop end-user CBDM tools in the certainty of a guaranteed market for the product by a due date – thus solving the “chicken and egg” conundrum noted above. Note that, although similar, or even identical, targets would be used with either approach, with CBDM it would be necessary to model user deployment of blinds etc since direct (and indirect) sunlight now figures in the evaluation. I hope that the issues raised here will be progressed in wider discussions within the daylighting community and relevant stakeholders. This article is the first of a series in support of the activities of CEN TC 169/WG11. The second article “A Roadmap for Upgrading National/EU Standards for Daylight in Buildings” was presented at the 2013 CIE Midterm conference in Paris, France [10]. The proposal made to the CEN Technical Committee WG11 is described in detail in the third paper of the series [11]. The fourth examines the practicalities and pitfalls in eventually upgrading to full climate-based metrics [12]. It should be noted that the views expressed in this paper are those of the author alone.

Acknowledgements The author owes a debt of gratitude to colleagues who serve on WG11, and also to Paula Esquivias Fernández (University of Seville), for many thought-provoking discussions on the nature of daylight metrics.

References [1]

J. Mardaljevic, L. Heschong, and E. Lee. Daylight metrics and energy savings. Lighting Research and Technology, 41(3):261–283, 2009.

[2]

M. Andersen, J. Mardaljevic, and S. W. Lockley. A framework for predicting the non-visual effects of daylight – Part I: photobiologybased model. Lighting Research and Technology, 44(1):37–53, 03 2012.

[3]

L. Heschong. Daylighting and human performance. ASHRAE Journal, 44(6):65–67, 2002.

[4]

J. A. Love. The evolution of performance indicators for the evaluation of daylighting systems. Industry Applications Society Annual Meeting, 1992., Conference Record of the 1992 IEEE, pages 1830–1836 vol.2, 1992.

[5]

P. Moon and D. E. Spencer. Illuminations from a non-uniform sky. Illum. Eng., 37:707–726, 1942.

[6]

Estidama. Pearls Design System - New Buildings Method. Abu Dhabi Urban Planning Council, 2009.

[7]

BB90. Lighting design for schools. Building Bulletin 90 (2008 revision), Department of Education and Skills, School Building and Design, London 2008.

[8]

D. Enarun and P. Littlefair. Luminance models for overcast skies: Assessment using measured data. Lighting Research and Technology, 27(1):53–58, 1995.

[9] D. R. G. Hunt. Improved daylight data for predicting energy savings from photoelectric controls. Lighting Research and Technology, 11(1):9–23, 1979. [10] J. Mardaljevic and J. Christoffersen. A Roadmap for Upgrading National/EU Standards for Daylight in Buildings. CIE Midterm conference – Towards a new century of Light, Paris, France 12-19 April, 2013. [11 J. Mardaljevic, J. Christoffersen, and P. Raynham. A Proposal for a European Standard for Daylight in Buildings. Lux Europa, Krakow, Poland, 17–19 September, 2013 (Abstract accepted, draft under review). [12] J. Mardaljevic. Daylight design, simulation and compliance for solar building envelopes. Energy Forum – Advanced Building Skins, Bressanone, Italy, 5–6 November, 2013.

SLL Advert 2013:Layout 1

09/10/2013

10:38

Page 1

Join a dynamic international community The Society of Light and Lighting EHQHÂ¿WV RI PHPEHUVKLS Â&#x2021; 1HWZRUN ZLWK RYHU OLJKWLQJ SURIHVVLRQDOV GHVLJQHUV UHVHDUFKHUV VWXGHQWV DQG PDQXIDFWXUHUV Â&#x2021; 3URIHVVLRQDO UHFRJQLWLRQ Â&#x2021; )UHH DFFHVV WR FLEVHNQRZOHGJHSRUWDO FR XN LQFOXGLQJ DOO 6// /LJKWLQJ *XLGHV DQG /LJKWLQJ 5HVHDUFK 7HFKQRORJ\ Â&#x2021; %L PRQWKO\ 6// QHZVOHWWHU ZLWK WKH ODWHVW LQGXVWU\ XSGDWHV Â&#x2021; )UHH VXEVFULSWLRQ WR WKH &,%6( -RXUQDO DQG DSS Â&#x2021; 8S WR PHPEHUV GLVFRXQW IRU 6// DQG &,%6( HYHQWV

8SFRPLQJ HYHQW 6// 0DVWHUFODVV 0DUFK 7LWDQLF %HOIDVW %ULQJLQJ \RX WKH H[SHUWV LQ WKH Â¿HOGV RI FRQWUROV ODPSV DQG OXPLQDLUHV OHJLVODWLRQ DQG GD\OLJKW

)UHH PHPEHUVKLS IRU IXOO WLPH VWXGHQWV -RLQ WRGD\ DV DQ DIÂ¿OLDWH PHPEHU IRU Â&#x2026;

ZZZ VOO RUJ XN )ROORZ XV #6//

LEDs are the panacea â&#x20AC;&#x201C; and other fairy tales

James Thomas Duff and Peter Whitty, Arup james.duff@arup.com

Building Servicesnews

CIBSE Sustainable Awards 2013

Abstract Recently, there has been an explosive growth in

Introduction

the popularity of LED luminaires. Like all markets

Once upon a time there were three lighting designers with a small practice in Dublin. They weren’t rich, but they worked hard and made a living by offering their clients up-to-date, innovative and honest advice about all aspects of lighting. On the back of this, they had built a commendable reputation and thoroughly enjoyed their occupations. However, in the present day, they approach their work with angst and fear. You see, in recent years, they have endorsed LED products with much excitement, but failed to properly interrogate manufacturer claims. The consequences of their actions have begun to surround them. Client complaints – failed fittings, insufficient light, noticeable colour shift, rapid lumen depreciation, extreme flicker, to name but a few. In short, their clients’ complaints highlight broken promises and demonstrate disappointment. For our designers, this is unfamiliar territory. They are competent, experienced and have always made decisions with their clients’ best interests in mind, yet they suddenly find themselves the focal point of many justifiable complaints. Where did it all go wrong? It’s obvious; they took a step into the unknown and placed their reputation in the hands of others, purely to keep up with the Joneses.

experiencing such expansion, it has attracted many new manufacturers. Some of these have a reputation to uphold, but many do not. As such, it can be difficult for designers to differentiate good quality products from bad quality products. This paper explores a set of standardised test methods and performance criteria associated with LED luminaires. Moreover, it proposes a set of questions for discussion within the lighting community. These questions would be used to interrogate manufacturer data and aid with the determination of product quality.

Key Words: LED, specification, questions, manufacturers.

This paper examines reasons behind client complaints, discusses quality criteria associated with LED luminaires and ultimately, proposes a draft set of questions that designers should ask manufacturers of LED luminaires to help differentiate good quality from poor quality products. Construction and operation of LED luminaires is explained in great detail elsewhere[1] – [10] and will not be covered here.

The growth of LED Boyce suggests[11] that the growth of LEDs has happened for three reasons. The first is the immense quantity of money invested in LEDs by organisations and the consequent rapid development in their capabilities. The second has been the enthusiasm of regulators who see LEDs as the ultimate replacement for incandescents. The third is fashion. At present, opting for LEDs is considered progressive and enlightened. As a result of these factors, the market is now saturated with new, and unfamiliar, LED products. This raises an issue for designers: how can they distinguish good equipment from bad?

A level playing field Traditionally, not many standards have been in place to aid designers in the selection of LED products, but in recent years, this has improved somewhat with the introduction of IESNA LM79-08, IES Approved Method for the Electrical and Photometric Measurement of Solid-State Lighting Products [12] and IESNA LM-80-08, IES approved Method: Measuring Lumen Maintenance of Light Emitting Diode Light Sources [13]. Both of these test methods allow manufacturers to have their products tested in an independent laboratory, to a standard set of testing procedures

LEDs are the panacea – and other fairy tales

that have been devised to examine particular qualities associated with LED products. The International Electrotechnical Commission (IEC) has also produced two publicly-available standards (PAS) that detail performance requirements for LED products; IEC/PAS 62717 Performance requirements – LED modules for general lighting [14] and IEC/PAS 62722 Performance requirements – LED luminaires for general lighting [15]. These documents deal with LED modules and LED luminaires separately, but both bring clarity to the manner in which LED data should be measured and presented by introducing a universal set of quality criteria. Details of the quality criteria should be displayed on LED product datasheets. This information offers designers an equal platform where they can evaluate products from different manufacturers. The quality criteria described within the IEC/PAS documents are: input power, luminous flux, luminaire efficacy, luminous intensity distribution, correlated colour temperature (CCT), colour rendering index (CRI), chromaticity co-ordinate values (initial and maintained), lumen maintenance code, photometric code, life in hours and the associated rated lumen maintenance (Lx), failure fraction (Fy), ambient temperature (tq), power factor and drive current. Some of these will be familiar terms, as they are commonly employed to describe features of traditional source luminaires. Nonetheless, a number of these will be less recognisable or their significance may be greater with LED sources. These are explained and briefly discussed in the following sections.

Colour rendering index (CRI) Despite the prominence of CRI, it has several shortcomings and problems[16]–[19]. Two luminaires that have the same white colour appearance may be the result of various blends of wavelengths[4][20]. Consequently, a given material may project dissimilar appearances, since the material surface will reflect the constituent wavelengths by varying extents – i.e. its colour appearance will change when it is exposed to one or other luminaire. The Commission Internationale de l’Eclairage (CIE) is currently in the process of revising the metric and recommends that it is not suitable for quantifying the colour rendering capabilities of white LED light sources[21]. As such, and for other reasons stated later in this paper, samples of LED luminaires should always be viewed before specification. The IEC/PAS documents specify CRI as a code and this code is given as indicated in Table 1.

Table 1 – IEC/PAS colour rendering index codes Code

CRI range

Colour rendering properties

57 – 66

Poor

67 – 76

Moderate

77 – 86

Good

87 – 100

Excellent

centre of the ellipse[23] – [26]. Therefore, the contour of the ellipse represents the just-noticeable differences of chromaticity. The scale of a single MacAdam ellipse is extremely small and, as such, they are usually scaled up to a larger size. This is represented by a 3-step, 5-step or 7-step MacAdam ellipse, where a 3-step is three times greater than a standard MacAdam ellipse, and so forth. When products are tested in accordance with the IESNA test methods quoted previously, initial and maintained chromaticity coordinates are measured, where the maintained value is recorded at 25% of rated life, but a maximum of 6,000 hours. The relevant IEC/PAS code can be obtained using Table 2. Table 2 – IEC/PAS initial and maintained colour variation codes Size of MacAdam ellipse, centred on the colour target

Colour rendering properties Initial

Maintained

3-step ellipse

5-step ellipse

7-step ellipse

> 7-step ellipse

Lumen maintenance code The life of LED luminaires is typically very long. It is therefore impractical to measure the actual reduction in lumen output over entire luminaire lifetime. Instead, validation of lumen maintenance at finite test times is specified by the IEC/PAS documents[14][15]. The maintained luminous flux is measured at 25% of rated lifetime up to a maximum of 6,000 hours. Therefore, the code numbers assigned do not imply a prediction of achievable life; they simply declare the quantity of luminous flux still emitted at 25% of rated life or 6,000 hours. The IEC/PAS code can be obtained using Table 3. Table 3 – IEC/PAS Lumen Maintenance Codes Code

Lumen maintenance (%)

≥ 90

≥ 80

≥ 70

In order to generate a lifetime claim, an extrapolation of test data is necessary. At present, the IEC has no standardised methodology for the extrapolation of lumen maintenance data[27]. However, if a product has been tested in accordance with IESNA LM-80-08[13], a standard extrapolation method IESNA TM-21-11, IES Approved Method: Making Useful LED Lifetime Projections is used to make extrapolated predictions[28]. This offers a standardised platform from which a designer can independently compare lifetime claims from various manufacturers.

Rated initial and maintained chromaticity coordinates In the study of colour vision, MacAdam ellipses refer to the region on a chromaticity diagram that contains all colours which are indistinguishable to the average human eye, from the colour at the

Photometric code The IEC/PAS photometric code is a six-digit code that displays the important “quality of light” parameters of an LED luminaire[14][15].

CIBSE Sustainable Awards 2013

Ambient temperature (tq) Initial colour variation

Colour rendering index

Code XXX / XXX

Correlated colour temperature

Maintained luminous flux

Maintained colour variation

Figure 1 â&#x20AC;&#x201C; IEC/PAS photometric code breakdown

It states initial CRI, initial CCT, initial and maintained colour coordinates and lumen depreciation code. Figure 1 above indicates in more detail what each digit signifies. As an example, consider an LED luminaire with the following properties and the corresponding photometric code indicated in brackets; initial CRI value of 83 (code 8), initial CCT value of 4000K (code 40), initial spread of chromaticity coordinates within a 3-step MacAdam ellipse (code 3), maintained spread of chromaticity coordinates within a 5-step MacAdam ellipse (code 5) and maintained luminous flux of 92% (code 9). The photometric code for this luminaire would be 840/359.

Rated life and associated lumen maintenance (Lx) Rated life is the length of time during which an LED module provides more than the claimed percentage of the initial luminous flux. This is stated in hours and represented in terms of Lx, where x is the percentage of lumen depreciation and should always be published along with the failure fraction (Fy). The recommended series of values for x is 70, 80 and 90[14][15]. It is important to note the limitations of LED lumen maintenance measurements. As briefly discussed previously, many LED manufacturers use test results provided by IESNA LM-80-08 to obtain lumen maintenance thresholds of LEDs, but there is a distinct disconnection between the results found by the LED manufacturer and the results for an LED luminaire. The performance of an LED module will vary within different luminaire housings and with varying thermal management systems[27]. In practice, LED manufacturers test their products to IESNA LM80-08 and then apply extrapolation procedures as detailed in IESNA TM-21-11 to arrive at L90, L70 or L50 figures[27][29]. Luminaire manufacturers then translate these curves into LED luminaire-specific curves. There are two constraints with interpreting data in this manner. The first and most crucial is that there is no validated method for translating the lumen maintenance curve for an individual LED into a curve for an LED luminaire. The second is that catastrophic failure of individual LEDs and failure of other critical luminaire components is not considered.

Failure fraction (Fy) Failure fraction is the percentage of a number of LED modules of the same type at their rated life that have failed. The failure fraction expresses the combined effect of all components of a module including mechanical, as far as the light output is concerned. The recommended series of values for y is 10 and 50[14][15].

For a given performance claim, the ambient temperature is a fixed value. The ambient temperature is important as it affects junction temperature, which affects lumen output, efficacy and luminaire life[1][5][6]. It is possible to specify performance claims at different temperatures and reputable manufacturers should be able to provide this. Temperature should be given in degrees Celsius[14][15].

Drive current Drive current affects LED junction temperature, which affects lumen output, luminous efficacy and luminaire life[1][5][6]. Typical drive current values are between 350mA and 700mA, but this can be higher if the thermal management of the luminaire is good enough to keep the junction temperature below critical failure temperature. In general, as drive current increases, lumen output will increase, efficacy will decrease, as will luminaire life. Reputable manufacturers will be able to provide a graphic demonstrating how life and efficacy vary with alternative drive currents.

Interrogating manufacturer data While the aforementioned documents offer the prospect of equality between manufacturers, it will be evident to those working in lighting that this information is not yet freely available on data sheets as the IEC/PAS standards advocate it should be. Boyce has suggested[11] that a valuable contribution to the lighting community would be a set of simple questions to ask the LED supplier, whereby any supplier that is unwilling, or unable, to answer these questions should be treated with caution. This section proposes a draft set of questions that could be put to manufacturers of LED luminaires in order to determine the quality of the product that they are supplying. The questions are derived partly from the IEC/PAS performance requirements and partly from the experience of the authors. They are worded in a manner that indicates to suppliers and manufacturers exactly what is required from their response. In the paragraph following this, a short explanation of expected responses is detailed.

Questions for manufacturers and suppliers of LED luminaires i.

Is this LED luminaire tested in accordance with IESNA LM-7908, IES Approved Method for the Electrical and Photometric Measurement of Solid-State Lighting Products? If so, please provide a results certificate.

ii. Are the LEDs within this luminaire tested in accordance with IESNA LM-80-08, IES approved Method: Measuring Lumen Maintenance of Light Emitting Diode Light Sources? If so, please provide a results certificate. iii. Is luminaire life extrapolated in accordance with IESNA TM-21-11, IES Approved Method: Making Useful LED Lifetime Projections? If it is not, is it estimated in accordance with any other standard procedure or extrapolation method? iv. Is the data in your product specification sheet presented in accordance with IEC/PAS 62717 Performance requirements,

LEDs are the panacea – and other fairy tales

LED modules for general lighting’ and ‘IEC/PAS 62722 Performance requirements, LED luminaires for general lighting? v. Who supplies the LEDs within your LED luminaires? vi. What is the rated Luminaire Lifetime? Provide an answer in terms of hours at LxxFxx, where Lxx represents parametric failure rate and Fxx represents the catastrophic failure rate. vii. State the luminaire photometric code, as defined within IEC/PAS 62722 and IEC/PAS 62717, where the six-digit code displays the important “quality of light” parameters of an LED luminaire. It should state initial CRI, initial CCT, initial and maintained colour variation and lumen depreciation code. Please provide a lumen depreciation curve. viii. What is the driver power factor and is the driver replaceable? ix. What is the total luminaire wattage, including control gear? x. What is the ambient temperature for which the luminaire performance is rated? Please provide information on how at least three ambient temperatures will affect the performance and lifetime of the LED luminaire. xi. What is the initial luminaire lumen output, for the specified driver current? If the luminaire is to be driven at a current that is not the standard, please provide information on how this will affect luminaire performance and life. xii. What length of time is the complete luminaire warrantied for? Please supply a copy of the warranty.

Acceptable responses? Now that we have a grasp of the questions we should be asking, what are the responses that we will consider acceptable or unacceptable? The following is merely the opinion of the authors, but has been formed through extensive discussion with noteworthy lighting designers, engineers, through social media, internet forums and interaction with very reputable manufacturers of LED luminaires. All LED modules and LED luminaires should be tested in accordance with IESNA testing methods and have product information displayed in accordance with IEC/PAS requirements. The cost of these tests is relatively low, so there is no justifiable reason why manufacturers would not test to IESNA standards and present data in accordance with IES/PAS requirements. It is preferable that all luminaire lifetimes are extrapolated in accordance with IESNA TM21-11, but this will not always be the case, particularly with manufacturers that do not retail in the USA. Some reputable suppliers of LED chips and modules are: Cree, Sharp, Philips, Osram, Epistar/Intermolecular, Nichia, Xicato, Citizen, Bridgelux and ASM Pacific, but this list is not exhaustive. LED luminaire manufacturers that refuse to disclose the manufacturer of the LED chips within their luminaires should be dismissed quickly. Luminaire lifetime with good quality products for general lighting is typically L70F10 for 50,000 hours. This may shorten or be stated as L70F50 for decorative or some architectural type luminaires and may lengthen for luminaires in cooler ambient temperatures. The first half (CRI and CCT) of the IEC/PAS photometric code will differ on a project-specific basis, but the

latter half, for good quality products, should be within a 3-step MacAdam ellipse initially and within a 3-step Macadam ellipse through lumen maintenance (for white light), combined with a lumen maintenance of Code 9 or better. Again, this may shift for decorative, external and architectural applications and products of lesser quality will generally show larger colour shift and greater lumen depreciation. This may be acceptable for some applications and not for others, but at least the designer and client will be aware of probable issues. Care should be taken with replaceable drivers, as by law and in accordance with European Commission Electromagnetic Compatibility Directive 2004/108/EC and all associated modifications[30], they cannot simply be replaced on-site in a plug and play manner. This is currently not enforced, but if an LED module or driver fails, the entire fitting should be replaced or a single component replaced and the fitting re-tested with the new component installed[30]. Combine this with the fragile nature of electronic components within an LED luminaire and the wide range of replacements available, and the authors believe that replaceable drivers should be avoided when possible as they are likely to cause unforeseen compatibility issues. The quoted total wattage should be given inclusive of control gear losses. Driver current affects lumen output, efficacy and luminaire life. Typically, increased drive currents will increase lumen output, but reduce efficacy and decrease life[1][5]. Similarly, increased ambient temperatures will generally decrease lumen output and luminaire life[1][5]. Reputable manufacturers will be able to supply information about how drive current and ambient temperature affect all of these parameters. For most of the better manufacturers of LED luminaires, their standard product warranty will be five years. It should be noted that some of the best LED manufacturers now offer a 10-year warranty, as standard, on their external LED luminaires, inclusive of all luminaire components. A copy of all product warranties should be obtained and examined, as they can often be misleading. Ensure that the warranty states very clearly what parameters will need to have drifted and how much they will need to have drifted by before the luminaire is replaced. Also be sure that all components are covered in the event of failure, i.e. the warranty should cover driver, heat sink, LED module, housing, etc.

The role of mock-ups Despite the quality criteria outlined, there are some aspects of LED quality which are best assessed using first-hand experience. Flicker and dimming are not discussed above. A combination of percentage flicker and flicker index can be used to estimate the stroboscopic effect of a luminaire[31]. Dimming is frequently specified by the protocol selected, but a common complaint is that LED luminaires will not dim to 1% of full lumen output. Both of these issues are effortlessly solved with a simple, real life mock-up. Aside from resolving these issues, when multiple luminaires are included in the mock-up, the designer and client can obtain personal experience of initial colour variation in the luminaire batch and compare this with colour variation between surrounding

CIBSE Sustainable Awards 2013

traditional source luminaires. This should leave the client and designer with few surprises when the product is finally installed on site.

[17] Scanda J. Colorimetry: Understanding the CIE System. New York: John Wiley and Sons, Inc., 2007, pp. 25–78.

Conclusion

[19] Davis W, Ohno Y. Toward an improved colour rendering metric. Proceedings of SPIE 2005; 5941: 59411G.

This paper has shown how fair comparison between LED products is possible using internationally-recognised test methods and performance standards. It has also applied the experience of the authors and opinions of many within the lighting community to devise a draft set of questions and acceptable responses that can be put to manufacturers of LED luminaires to assist with determining the quality of an LED product. This list is not intended to be definitive, but rather to stimulate discussion within the lighting community and familiarise readers with what is required to differentiate between good and bad quality LED products.

References [1] Schubert, EF., Light-Emitting Diodes 2nd Edition. Cambridge University Press. 2006. ASIN B0087IC9FG. [2] Stockman, SA. Light-emitting Diodes: Research, Manufacturing, And Applications Ix; Ed. by Steve A. Stockman. Proceeding of SPIE. Society of Photo Optical. 2005. [3] Hall, JT. And Koskinen AO. Light-Emitting Diodes and Optoelectronics: New Research (Electrical Engineering Developments). Nova Science Pub Inc. 2012. ISBN 1621004481. [4] Illuminating Engineering Society of North America. The IESNA Lighting Handbook, 10th Edition, New York: IESNA. 2012. [5] Willardson, R., Weber, E., Stringfellow, G., and Crawford, M. High Brightness Light Emitting Diodes. Academic Press. ISBN 9780127521565. [6] Held, G., Introduction to Lighting Emitting Diode Technology and Applications. Auerbach Publications. 2012. ISBN 9781420076639. [7] Kitai, A., Principles of Solar Cells, LEDs and Diodes: The Role of the PN Junction. John Wiley & Sons. 2011. ISBN 9781119974543. [8] Markoc, H,. Nitride Semiconductor Devices: Fundamentals and Applications. John Wiley & Sons. 2013. ISBN 9783527649006. [9] Moram, MA., Light-emitting diodes and their applications in energy saving lighting. Proceedings of the IEC – Energy, Volume 164, Issue 1, January 2011, pages 17 – 24. [10] Seong, TE., Han, J., Amano, H., and Morkoc, H., III-Nitride Based Light Emitting Diodes and Applications (Topics in Applied Physics). Springer Link Publications. 2013. ISBN 9400758626. [11] Boyce P. R. Editorial: LEDs are the answer, now what’s the question? Lighting Research and Technology June 2013 45: 265. [12] The Illuminating Society of North America. LM-79-08, IES Approved Method for the Electrical and Photometric Measurement of Solid-State Lighting Products. ISBN: 978-0-87995-226-6. [13] The Illuminating Society of North America. LM-80-08, IES approved Method: Measuring Lumen Maintenance of Light Emitting Diode Light Sources. ISBN: 978-0-87995-227-3. [14] International Electrotechnical Commission. Publically Available Standard 62722 Performance requirements – LED luminaires for general lighting. ISBN 978-2-88912-567-8. [15] International Electrotechnical Commission. Publically Available Standard 62717 Performance requirements – LED modules for general lighting. ISBN 978-2-88912-476-3. [16] Rea MS. The IESNA Lighting Handbook: Reference and Application, 9th Edition. New York: Illuminating Engineering Society of North America, 2000.

[18] Guo X, Houser KW. A review of colour rendering indices and their application to commercial light sources. Lighting Research and Technology 2004; 36 (3): 183–197.

[20] Society of Light and Lighting, The SLL Lighting Handbook, 2009, London; Chartered Institute of Building Services Engineers. [21] Commission Internationale de l’Eclairage. Colour Rendering of White LED Light Sources. CIE Publication 177, Vienna: CIE, 2007. [22] The Society of Light and Lighting (SLL). 2009. The SLL Code for Lighting. ISBN-978-906846-07-7. London: Chartered Institute of Building Services Engineers. [23] MacAdam, DL., "Visual sensitivities to color differences in daylight". 1942. Journal of the Optical Society of America, 32 (5): 247–274. [24] Kühni, RG., "Historical Development of Color Space and Color Difference Formulas". Color Space and Its Divisions. 2003. New York: Wiley. ISBN 978-0471-32670-0. [25] Judd, Deane B. (July 1939). "Specification of Color Tolerances at the National Bureau of Standards". The American Journal of Psychology (The American Journal of Psychology, Vol. 52, No. 3) 52 (3): 418–428. [26] Wright, William David; Pitt, F.H.G. (May 1934). "Hue-discrimination in normal colour-vision". Proceedings of the Physical Society 46 (3): 459–473. [27] CELMA. Why standardisation of performance criteria for LED luminaires is important. Federation of National Manufacturers Associations for Luminaires and Electrotechnical Components for Luminaires in the European Union. 2011. [28] The Illuminating Society of North America. TM-21-11, IES Approved Method: Making Useful LED Lifetime Projections. ISBN: 978-0-87995-227-3. [29] Lighting Industry Liaison Group, A Guide to the Specification of LED Lighting Products. London 2012. [30] Implementing Directive 2004/108/EC of the European Parliament and of the Council on the approximation of the laws of the Member States relating to electromagnetic compatibility and repealing Directive. Official Journal of the European Union. Commission Regulation 89/336/EEC. 2004. [31] United States Department of Energy. Solid State Technology Fact Sheet – Flicker. Gateway Program. 2013.

Wind resource in the urban environment

Jonathan Blackledge Derek Kearney* Eamonn Murphy *Corresponding author:

derek.kearney@dit.ie

Building Servicesnews

CIBSE Sustainable Awards 2013

Abstract Renewable energy technologies, such as wind

1. Introduction

turbines, have to be considered for new building

The World Meteorological Organisation (WMO) Commission for Instruments and Methods of Observation recognised the need to include in the 1996 WMO Guide to Instruments and Methods of Observation, WMO-No. 8, a new chapter on Urban Observations (Oke, 2006). The foreword to the Report states that “…the realities for those faced with the establishment of a meteorological station at an urban site where application of standard siting is often either impossible or nonsensical.” (Oke, 2006). The driver for developing the guide is the ever-increasing demand for wind data from urban sites. An example of this is the European Communities Energy Performance of Buildings Directive SI 666 (2006), which came into force on 1st January 2007. Part 2 states that: “A person who commissions the construction of a large new building shall ensure, before work commences on its construction, that due consideration has been given to the technical, environmental and economic feasibility of installing alternative energy systems in the proposed large building, and that the use of such systems has been taken into account, as far as practicable, in the design of that building.”The alternative energy system is further defined as “… decentralised energy supply systems based on renewable energy …”. So it is reasonable to assume that micro-wind turbines should be part of the design considerations in any future buildings. Also, natural ventilation systems can help to reduce the energy rating of buildings.

over 1000m2 under the Energy Performance of Buildings Directive (2002). Accurate assessment of the wind resource is a key component in the success of a wind installation. Designers, planners and architects also need wind data from urban areas to support low-energy building design, natural ventilation, air quality, pollution control, insurance and wind engineering. Over the last six years instrumentation has been installed at the Dublin Institute of Technology (DIT) in two separate locations to monitor the wind. The data has shown that the wind resource will vary quite considerably on a given site and this is due to local variations in topography, and other factors associated with wind and turbulence in the built environment. Difficulties were encountered in measuring the wind and turbulence on site. IEC 61400-12-1: 2005 states that “... analytical tools (anemometers presently available) offer little help in identifying the impact of these variables, and experimental methods encounter equally-serious difficulties.” The practical experience of measuring wind in the urban environment informed the development of a prototype anemometer that may be capable of digitally mapping accurate real-time three-dimensional data on wind speed, wind direction and, uniquely in the field of wind instrumentation, wind turbulence.

Key Words: Wind, turbulence, natural ventilation, micro-wind turbines, three-dimensional anemometer.

Figure 1: Renewable energy plant installed on DIT Kevin St Campus roof as part of Dublin Energy Lab (DEL) experimentation

The objective of this research is to assess the variation in wind resource across two locations on the Dublin Institute of Technology (DIT) campus, and compare it with data taken from the National Meteorological Service in Ireland (Met Éireann). Cup anemometers have been installed on the two sites to monitor the wind. This data was used to provide an example of wind resource in the urban environment and to indicate the inherent difficulties and pitfalls associated with engineers and designers trying to assess the wind resource in the urban environment. See Figure 1 for an example of the research being carried out by members of the Dublin Energy Lab. (DEL).

Wind resource in the urban environment

Lessons learned from this assessment have informed the design of a new three-dimensional wind measurement instrument called a Metometer. The initial outcomes from a year of field trials on the three-dimensional anemometer have yielded some very promising results. The progress of initial developments is explained and industrial partners are being sought for the project.

2. Methodology The European wind energy resource map indicates that Ireland and Scotland enjoy the highest wind resource in Europe (Gardner, Garrad et al., 2009). The European map indicates a uniform wind speed at ground level inland. In reality though there are a number of different factors that can effect wind at a particular location such as obstruction by buildings or trees, the nature of the terrain, and deflection by nearby mountains or hills (Met-Éireann, 2010). An example of this is the rather low frequency of southerly winds at Dublin Airport and this is due to the sheltering effect of the mountains to the south. Another example of local topography causing variations in the wind speed is Leinster where average annual wind speeds range from 3 m/s in parts of south Leinster to over 8 m/s in the extreme north, which is approximately 100 km away. On average there are less than two days with gales with wind speeds above 17 m/s each year at some inland places like Kilkenny but more than 50 days a year at northern coastal locations such as Malin Head. The uninterrupted wind flow from the Atlantic makes the north and west coasts of Ireland two of the windiest areas in Europe. The local variations in wind have two principal causes. One is friction with the earth’s surface, which can be extended as far as flow disturbances caused by topographical features such as hills and mountains. The second is “thermal effects”, which can cause air masses to move vertically as a result of variations in temperature (Burton et al., 2001). As the height above the ground increases, the effect of earth’s surfaces weaken, so that by 600m above the highest local obstruction the wind is generally free from surface influences. Here the wind can be considered to be driven by largescale pressure differences and the rotation of the earth, and this air flow is known as the geostrophic wind. Below this level where the effects of the earth’s surface can be felt is known as the boundary layer.

2.1 Boundary layer The ground surface has the effect of reducing the speed of the wind and this is because of the drag. The level of drag will vary depending on the surface roughness and there are many charts of the roughness factor associated with various terrain. The drag caused by the roughness is transmitted to the wind at higher levels by the action of turbulent stresses (Best, Brown et al., 2008). The characteristics of the wind flow in the urban canopy layer are markedly different to the roughness layer (Ricciardelli and Polimeno, 2006). In the urban canopy layer the flow is influenced more by local geometry than by energy transfer between the different layers. The WMO, in its report on the Initial Guidance to Obtain Representative Meteorological Observations at Urban Sites (Oke,

Figure 2: Schematic of climatic scales and vertical layers found in urban areas. PBL – planetary boundary layer, UBL – urban boundary layer, UCL – urban canopy layer (Oke, 2006)

2006), turned its attention to the measurement of the wind resource within the boundary layer. Of particular interest to the WMO is the Urban Canopy Layer (UCL) which is beneath roof level and directly above it. The difficulty identified by the WMO is that most developed sites make it impossible for a weather station installed in an urban environment to conform to the standard installation and site location guidelines in the Guide to Meteorological Instruments and Methods of Observation (WMO, 2008). Figure 2 gives an indication of the complex nature of wind flow around buildings and this is further complicated by proximity to other buildings. The turbulence encountered in the urban environment adversely affects the performance of wind turbines in the urban environment. The existing cup anemometer, and wind vane indicator, does not accurately convey the level of turbulence present on a site (Hölling, Schulte et al., 2007). Experience from the micro-wind turbine installed in the Dublin Institute of Technology (DIT), Church Lane, Kevin Street has shown that the cup anemometer will spin quickly when the turbine is not moving and also there are occasions when the turbine will rotate even when the cup anemometer is not moving. In a series of articles for Renewable Energy Focus, Holdsworth (2009) identified a number of key areas where further research into micro-wind technologies is required. Working as a consultant in this area he identified a number of major impediments to the development of the urban wind industry. Holdsworth draws the distinction between wind at a height of 100 metres and wind closer to the ground in the urban environment. At a height of 100 metres wind speed and direction will be the same over a large area, whereas closer to the ground, as shown in Figure 2, the pattern changes due to resistance the wind meets from terrain roughness. Wind is also affected by the shape, height and relationships that buildings have with each other; the impediments of parks and streets; and the creation of dead-zones that alter according to higher-level wind flows. According to Holdsworth, what the urban wind industry needs at this point is a much more thorough understanding of the physics of wind.

CIBSE Sustainable Awards 2013

Figure 4: Monthly average of wind speed data from Church Lane, Focas Institute and Met Éireann weather stations

Figure 3: Two-dimensional flow around a building with flow normal to the upwind face (a) stream lines and flow zones; A – undisturbed; B – displacement; C – cavity; D – wake; and (b) flow, and vortex structures (Figure 3 courtesy Oke, 2006)

The reason why Holdsworth (2009) has identified this as such a problem is that a relatively small difference in average wind speed results in a big difference in the energy output of a turbine. Also, the wind turbine has to have the capability to adapt to the wind regime that occurs within the micro or meso climatic context of the building, or group of buildings, where it is installed, if it is to be effective. This industry is only in the early stages of development and there has been what Holdsworth describes as a “rush to please” in the urban wind industry, resulting in failures to achieve the capacity promised due to ignorance of the way wind energy behaves in an urban landscape. This has been reflected in the very poor results of recent trials in the United Kingdom (UK), where capacity factors as low as 1.5% have been recorded (Encraft, 2009). The poor power output of turbines has put the spotlight on the measurement of the power input … wind.

3. Wind resource measured at the Dublin Institute of Technology Wind data was taken from anemometers installed at the DIT over a number of years. The sample period for this paper is January to June 2009 and the sites chosen were the car park in Church Lane and the roof of the Focas building. The car park is totally enclosed by buildings and trees with the slightly more open view to the east; the anemometer was mounted on a pole approximately six metres above ground. The rationale behind choosing the Church Lane car park was that the site roughly matched the location where small wind turbines had been installed on houses, shops, businesses, etc in the UK and Ireland. The Focas building is a four-storey building that is open to the wind from all directions, except the north-east. Data from these two buildings were compared with an average of Met Éireann data from Malin Head, Johnstown Castle, Valentia and Kilkenny Weather stations.

Figure 5: Daily average of wind speed data from Church Lane, Focas Institute and Met Éireann weather stations

Figure 6: Hourly averages of wind speed data from Church Lane, Focas Institute and Met Éireann weather stations

Figure 4 shows that over the six months there was a consistent marked difference between the sites. The more exposed Met Éireann weather stations recorded a higher value of wind when compared with the Focas building and the Church Lane car park. As the mounting height and exposure of the cup anemometers decreased there was a corresponding decrease in the wind speed measured. This pattern is repeated in the daily and hourly averages of the data as shown in Figure 5 and Figure 6. A more in-depth analysis of the data measured at the DIT, using the Levy Index, is contained in: Wind turbine Power Quality Estimation Using a Lévy Model for Wind Velocity Data (Blackledge, Coyle et al., 2011). The disappointingly low wind resource recorded is not the only consideration when evaluating the wind speed as turbulence also needs to be taken into consideration.

3.1 Turbulence Turbulence refers to fluctuations in wind speed in a time scale of less than ten minutes, with generally lower timescales for the urban environment. Burton et al., (2001) states that it is useful to consider wind as having seasonal and daily variations with turbulent fluctuations superimposed. Turbulence is generated mainly from

Wind resource in the urban environment

two causes — firstly, friction with the earth’s surface, which is flow disturbances caused by the topographical features; and secondly, thermal effects, which can cause air masses to move vertically as a result of variations in temperature. Turbulent flow is by its very nature chaotic … the flow velocity is very sensitive to perturbations and fluctuates wildly in time and in space. Turbulent flow contains swirling flow structures (eddies) with characteristic length, velocity and time scales which are spread over very wide ranges (Burden, 2008).

Anemometer

Turbulence in the wind is caused by dissipation of the wind’s kinetic energy into thermal energy and this occurs through the creation of progressively-smaller eddies (Manwell, McGowan et al., 2009). Turbulent wind generally has a very variable pattern over a short timeframe but it has a relatively constant average over longer time periods. This is why the statistical properties of turbulence are a common means of evaluating the effect of turbulence. There have been many definitions of turbulence but there is currently no universally-accepted definition. In 1937 Taylor and Von Karman gave the following definition: “Turbulence is an irregular motion which in general makes its appearance in fluids, gaseous or liquid when they flow past solid surfaces, or even when neighbouring streams of the same fluid flow past one another.” (cited in Hinze, 1976). So, from this definition a flow has to be irregular to be considered as turbulent. The National Renewable Energy Laboratory in America (Bailey and McDonald, 1997) state that “Wind turbulence is the rapid disturbances or irregularities in the wind speed, direction and vertical component.” The most common indicator of turbulence is the standard deviation (σ) of wind speed. When σ is normalised with the average wind speed it gives the Turbulence Intensity (TI), which gives an indication of a site’s turbulence. On this scale low levels are indicated by values less than, or equal to, 0.10; moderate levels to 0.25; and high levels greater than 0.25. TI is defined as: TI = σ where: V

σ = the standard deviation of wind speed; and V = the mean wind speed. (Bailey and McDonald, 1997) IEC 61400-12-1: 2005 is the only standard that considers the power performance of wind turbines. This standard requires a calculation of turbulence intensity. In an urban environment the turbulence intensity will be higher due to the local obstructions.

3.2 Turbulence intensity An anemometer has been installed on the Focas building in the DIT for the last number of years. It is installed below the roof level and subject to many obstacles as shown in Figure 7.

Figure 7: Anemometer mounted on bracket over PV array on roof of DIT Kevin St Campus renewable research facility

Table 1 – Data from the Focas building DIT for a 10 minute period on the 31st January 2009 Time

Wind Direction

Wind Speed

Humidity

Temp

Bar.

23:40

1.2

23:41

173

1.6

756.8

6.8

756.8

23:42

23:43

149

3.9

6.9

756.8

2.4

6.9

756.8

23:44 23:45

2.8

756.8

1.4

6.9

756.8

23:46

118

1.6

6.8

756.8

23:47

118

1.6

6.8

756.8

23:48

2.6

6.8

756.8

23:49

263

1.6

6.8

756.8

The turbulence intensity for the site can be calculated. TI = σ V

When considering the data from the Focas building the average value of wind speed for the 10-minute period is calculated. Average = 2.07 m/s The standard deviation of the wind speed from the average is: σ¬ = 0.79

The weather station on the Focas Building, DIT logs a number of parameters and a sample of one-minute averaged data is shown in Table 1.

The turbulence intensity for the site can then be calculated.

The standard deviation of the wind speed from the average can be calculated.

The turbulence intensity 0.385 measured at the Focas is high as the National Renewable Energy Laboratory (NREL) USA indicates that low levels of turbulence have values less than, or equal to 0.10; moderate levels to 0.25; and high levels greater than 0.25 (Bailey

∑ (x − x )

σ =

(n )

T .I . =

σ 0.841 = = 0.385 U 2.07

CIBSE Sustainable Awards 2013

and McDonald, 1997). The 0.385 measured at the Focas building is in line with the Warwick Wind Trials report which shows that the turbulence intensity is greater for the lower mounting positions and that it is increased by the presence of surrounding buildings (Encraft, 2009). This is due to the presence of aerodynamic friction and thermal gradients which are responsible for the creation of atmospheric turbulence (Cochran, 2002).

3.3 Wind charger To demonstrate the effect that turbulence can have on the performance of a micro-wind turbine, a domestic 220-watt wind charger was installed in the car park of Church Lane at a height of six metres, as shown in Figure 8.

Table 2 – Performance of wind turbine in Church Lane over a six-month period Jan

Feb

Mar

Apr

May

Jun

Overall

Average 1.3 Wind Speed (m/s) Average 0.45 Power Produced (W) Power 0.33 Produced (W) (kWhr)

0.63

1.09

0.84

1.11

0.98

0.17

0.12

0.17

0.26

0.22

0.23

0.13

0.089 0.127

0.193

0.164 1.033

produced in a given period, to the theoretical maximum possible. A reference wind turbine installed in an open field had a capacity factor of 10.3% and the best turbine in the trial had a capacity factor of 4.4%.

4. Three-dimensional anemometer – Metometer

Figure 8: 220W wind charger installed in Church Lane, DIT Kevin St Campus

The turbulence index according to IEC 61400-12-1: 2005 could not be calculated for this site as the data was not available in sufficient resolution. However, the turbulence index for this site would be significantly higher than the 0.385 measured on the Focas building nearby, due to the lower mounting height. Notwithstanding the fact that the turbine was installed in the location of high turbulence to match similar installations observed in other locations where the installers had complained of poor results, there was still general surprise at the low performance figures of the turbine (Encraft, 2009). The turbine installed could be seen constantly “hunting” for the direction of the available wind and therefore producing very little power. The power output for the turbine is shown in Table 2 with some correction for gaps in the data measured. The performance of the micro-wind turbine is comparable with international experience. The Warwick Wind Trials measured the output of 30 small wind turbines installed in the urban environment in the UK (Encraft, 2009). The average capacity factor for the trial was 1.7%, the capacity factor being the ratio of the actual energy

The problems identified by observing the instrumentation installed at the DIT was that there were quite clearly swirling winds with very fast fluctuations present on-site that were not represented by the data, or the turbulence model shown in IEC 61400-12-1: 2005. Also, the turbine spent most of the time hunting for the wind direction and as a result produced very little power. IEC 61400-121: 2005 states that “...identical wind turbines will yield different power at different sites even if the hub height wind speed and air density are the same. These other variables include turbulence fluctuations of wind speed (in three directions), the inclination of the flow vector relative to horizontal, scale of turbulence and shear of mean wind speed over the rotor. Presently, analytical tools offer little help in identification of the impact of these variables and experimental methods encounter equally-serious difficulties.” For the electrical power output side of a wind turbine there are instruments to measure all of the parameters such as current, voltage, harmonics, frequency, etc. present, yet there is no equivalent instrument for the wind industry in an urban or turbulent environment. The need for a new wind measurement instrument capable of measuring three-dimensional variations in the wind and turbulence was identified. The Metometer uses multiple Pitot tubes incorporated into a spherical design to provide simultaneous real-time data on wind speed, direction and turbulence. The sample and record frequency of the Metometer is up to 1,000Hz, and this is in three-dimensions. The three-dimensional anemometer has a wide range of potential applications that include: • Enhanced site evaluation, planning, development and monitoring; • Capability for wind farms and tidal energy farms; • Superior meteorological data collection and analysis; • Urban planning applications for civil and mechanical engineers;

Wind resource in the urban environment

• Wide measurement range – wind speeds from 0-250 m/sec can be measured with a certified level of accuracy to less than 0.05% Full Scale Output (FSO); • Ease of manufacturing – because the device is based on an innovation in the design concept – has no moving parts and can be built using accessible materials. Aso, it can be easily manufactured at scale. The stage of development of the Metometer is that a full-scale prototype device has been designed, built and tested. Initial test data has been collected which confirms the high degree of accuracy and frequency of sampling from the device. A patent application was filed in November 2012 by DIT. Currently data logging and interface software is being developed to support the device. DIT is seeking partners and collaborators to licence the technology, or to develop the prototype instrument further into a market-ready product.

5. Conclusion

Figure 9: Prototype three-dimensional anemometer, called a Metometer, in author’s back yard

• Improved data for decision-making in aviation to enhance efficiency and safety; • Road safety applications. The very high sampling rate of the device compares very well with other instruments currently on the market, such as cup anemometers and wind vanes, which have an average sampling rate of just 1 Hz. Sonic anemometers have a sampling rate in the range of 20Hz. The resolution of data provided by the Metometer is only limited by the frequency required by industry, and by the natural fluctuations of wind or fluid to be measured. Wind speeds from 0-250 m/sec can be measured with a certified level of accuracy to less than 0.05% Full Scale Output (FSO). Data from initial trials carried out over 12 months has proven the device to be robust, and it can be built to withstand practically all environments. The advantages of the Metometer include: • Improved data accuracy – greater accuracy of wind data, including fluid speed, fluid direction and turbulence;

Engineers and designers working in the built environment would need to make a very careful assessment of the site before considering any device such as a turbine or natural ventilation system that relies on the natural wind resource to operate. With an average wind speed that can be as low as 0.15 times the national average measured by Met Éireann, it can be seen that the built environment has a very significant effect on the wind resource. Even when the lower-than-average wind speeds in the built environment are taken into consideration, there is also the additional factor of turbulence. Current wind instrumentation, according to the WMO, is not up to the job of providing accurate data on the wind resource in the built environment. The Metometer has demonstrated this capability in the prototype device already built. Enterprise Ireland has provided nearly €300,000 in funding to further develop and test the prototype instrument.

Acknowledgements The authors acknowledge the support of Enterprise Ireland, Science Foundation Ireland, and the Dublin Energy Lab. The authors also wish to acknowledge the assistance of Thomas Shannon and Noel Masterson for their help with the software and hardware associated with the collection of data. Tim Oke of the World Meteorological Organisation, has kindly agreed to allow the use of two of his diagrams in this paper.

• 3-D Capability – improved data quality by measuring in three dimensions; • Sampling frequency – greater number of samples can be taken within a specified timeframe; • Robust design – device is engineered to be durable, and can be manufactured using a range of materials, including stainless steel; • Low maintenance – design has no moving parts, therefore maintenance is low, and the effects of rain, frost, snow, dust, or sunshine significantly reduced;

CIBSE Sustainable Awards 2013

References Bailey, B. H. & McDonald, S. L. (1997) Wind Resource Assessment Handbook. Fundamentals for Conducting a Successful Monitoring Program. National Renewable Energy Laboratory. Best, M., Brown, A., Clark, P., Hollis, D., Middleton, D., Rooney, G., Thomson, D. & Wilson, C. (2008) Small-Scale Wind Energy – Technical Report. In Lothingland, R. (Ed.) Urban Wind Energy Research Project. Exeter, Met Office - Carbon Trust. Blackledge, J., Coyle, E. & Kearney, D. (2011) Wind turbine Power Quality Estimation Using a Lévy Model for Wind Velocity Data. IEEE 10th International Conference on Environment and Electrical Engineering (EEEIC), 2011 Rome. Burden, T. (2008) The First Few Lectures in a First Course on Turbulence. Tony Burden’s Lecture Notes, Spring 2008. Burton, T., Sharpe, D., Jenkins, N. & Bossanyi, E. (2001) Wind Energy Handbook, Chichester, John Wiley & Sons, Ltd. . Cochran, B. C. (2002) The Influence of Atmospheric Turbulence on the Kinetic Energy Available During Small Wind Turbine Power Performance Testing. IEA Expert Meeting on: Power Performance of Small Wind Turbines Not Connected to The Grid. Soria, Spain, Cermak Peterka Petersen, Inc. Encraft (2009) Warwick Wind Trials. In Hailes, D. (Ed.) Warwick, Warwick District Council. Energy Performance of Buildings Directive 2006. SI 666. Ireland. Gardner, P., Garrad, A., Hansen, L. F., Jamieson, P., Morgan, C., Murray, F. & Tindal, A. (2009) Wind Energy – The Facts. Part 1: Technology. Intelligent Energy - Europe Programme European Wind Energy Association. Hinze, J. O. (1976) Turbulence, New York ; London, McGraw-Hill. Holdsworth B. (2009) Options for Micro-Wind Generation: Part 1 & 2. Renewable Energy Focus News. Hölling, M., Schulte, B., Barth, S. & Peinke, J. (2007) Sphere Anemometer - a Faster Alternative Solution to Cup Anemometry. Journal of Physics: Conference Series 75. IEC 61400-12-1: 2005 Wind Turbines. Power Performance Measurements of Electricity Producing Wind Turbines International Electrotechnical Commission. Manwell, J. F., McGowan, J. G. & Rogers, A. L. (2009) Wind Energy Explained – Theory, Design, and Application, Chichester, John Wiley & Sons. Met-Éireann (2010) Climate of Ireland - Wind. Dublin, Met Éireann. Oke, T. R. (2006) Initial Guidance to Obtain Representative Meteorological Observations at Urban Sites. Instruments and Observing Methods World Meteorological Organization. Ricciardelli, F. & Polimeno, S. (2006) Some characteristics of the wind flow in the lower Urban Boundary Layer. Journal of Wind Engineering and Industrial Aerodynamics, 94, 815-832. WMO (2008) Guide to Meteorological Instruments and Methods of Observation. Geneva, World Meteorological Organization.

The Small Wind Energy Estimation Tool (SWEET) â&#x20AC;&#x201C; a practical application for a complicated resource

Keith Sunderland* Thomas Woolmington Gerald Mills Jonathan Blackledge Michael Conlon *Corresponding author:

keith.sunderland@dit.ie

Building Servicesnews

Keith Sunderland SDAR paper:Layout 1

17/10/2013

09:33

Page 30

CIBSE Sustainable Awards 2013

Abstract Of the forms of renewable energy available, wind

incorporated into a practical tool developed in EXCEL,

energy is at the forefront of the European (and

namely the Small Wind Energy Estimation Tool

Irish) green initiative with wind farms supplying a

(SWEET). This tool is designed to assist engineers gain

significant proportion of electrical energy demand.

an intuitive appreciation of the limitations associated

Increasingly, this type of distributed generation (DG)

with this form of energy. It is only through an

represents a â&#x20AC;&#x153;paradigm shiftâ&#x20AC;? towards increased

understanding of such limitations that informed

decentralisation of energy supply. However, because

decisions can be made which ultimately facilitate

of the distances of most DG from urban areas where

more intelligent installations

demand is greatest, there is a loss of efficiency. One possible solution, placing smaller wind energy systems in urban areas, faces significant challenges. However, if a renewable solution to increasing energy demand is to be achieved, energy conversion systems in cities, where populations are concentrated, must be considered. That said, assessing the feasibility of small/micro wind energy systems within the built environment is still a major challenge. These systems are aerodynamically rough and heterogeneous surfaces create complex flows that disrupt the steady-state conditions ideal for the operation of small wind turbines. In particular, a considerable amount of uncertainty is attributable to the lack of understanding concerning how turbulence within urban environments affects turbine productivity. This paper addresses some of these issues by providing an improved understanding of the complexities associated with wind energy prediction. This research used detailed wind observations to model its turbulence characteristics. The data was obtained using a sonic anemometer that measures wind speed along three orthogonal axes to resolve the wind vector at a temporal resolution of 10Hz. That modelling emphasises the need for practical solutions by optimising standard meteorological observations of mean speeds, and associated standard deviations, to facilitate an improved appreciation of turbulence. The results of the modelling research are 30

Key Words: Small wind turbines, urban environments, turbulence, turbulence intensity, Gaussian and Rayleigh distributions.

The Small Wind Energy Estimation Tool (SWEET) – a practical application for a complicated resource

1. Introduction To produce electrical energy, wind turbines extract kinetic energy from moving air and convert it into mechanical energy, from which the turbine rotor derives electricity through the generator. Micro wind turbines can be either horizontal axis (HAWT) or vertical axis (VAWT) and are distinguished by blade diameter, cut-in/rated wind speed, and output power at rated wind speed. The two defining aspects of a wind turbine’s performance are the blade sweep area and the associated power curve for the turbine. The blade sweep area defines the amount of power that can be captured from the available wind while the power curve describes the turbine’s performance against varying wind speeds. The mechanical energy captured by the wind rotor is defined by the Betz constant in the equation describing mechanical power through wind energy:

(1)

where – Cp, is the power coefficient, defining how much wind power is captured and turned into mechanical power to subsequently generate electricity – ρair, is the mass density of air; – Arotor, is the rotor area ( .R2 where R is the length of the blades); – u is the wind speed. According to the Betz limit, the maximum possible conversion coefficient for any laminar kinetic process is 59.3%. However, there are underlying assumptions in this calculation that all the energy conversion is kinetic in nature and that the mass flow is nonturbulent. This Betz limit does not strictly hold true for a turbulent wind resource, which reduces the overall power-producing capability for a given technology. Also, as turbulence is very much site-specific, a viable metric needs to be used in order to quantify its intensity and its likely effects on a given turbine technology. In practice, losses due to (aerofoil) blade roughness, wake effects, hub loss and tip losses reduce the coefficient of performance to much lower values and, as such, typical values range from 0.1-0.35 for the majority of commercially-available small/micro turbines. If the wind is unsteady the energy conversion capability of the turbine is further degraded. Eqn (1) suggests that, in an ideal scenario, the power generated is proportional to the cube of wind speed (u3) but, for a number of reasons, this does not hold true for the majority of turbines. Firstly, the drag on most turbines is relatively static when compared to the wind speed at the height of the generator (hub-height); some of the newer techniques do use technologies such as furling of the blades or blade pitch control but these only respond within the operational range of the turbine, e.g. 3-15m/s. Secondly, the vast majority of grid-interfaced inverters function in a linear manner so that doubling the wind speed produces double the power. A block diagram description of how micro wind generation technologies connect in parallel with the distribution network (in Ireland) is

Figure 1: Block diagram of grid-tie logistics for small/micro wind systems contextualised in terms of the typical flow of power conversion

illustrated in Figure 1, contextualised with the flow of power conversion associated with these technologies. While the micro wind sector is still at an early stage of development, there is evidence of a growing market for micro wind systems[1]. From an Irish perspective, microgeneration technologies are identified in legislation as options for alternative energy supply. For example, the recently-published Building Regulations (Part L Amendment)[2] stipulate that for new installations, “a reasonable proportion of the energy consumption to meet the energy performance of a dwelling is provided by renewable energy sources”. In reality, micro-generation uptake in Ireland has been relatively poor, even though there is a technical and fiscal infrastructure in place. At the end of 2011 there was nearly 3MW of micro generation grid connected, representing an increase of 49.8% capacity compared to 2010 (1962.73kW[3]). While there is long way to go before the sector can make an impact on renewable energy targets, it is still the most embraced microgeneration technology of choice in Ireland with 79.2% of the 620 connections. The wind turbines considered in these statistics have installed capacities ranging from 500W up to 17kW, although under the micro generation connection[4] criteria, only technologies with ratings up to 6kW are considered as micro generation. The Irish Distribution Network Operator (DNO) ESB Networks – and the Commission for Energy Regulation (CER) – have a conservative attitude towards microgeneration in general. The UK, on the other hand, has a strong commitment to the embracement of power generation from small wind energy systems with 160.96 MW capacity across the sector[1]. Indeed, the UK Government has ambitious targets and aspires to have 2% of national electricity demand supplied from small scale source[5]. While wind power at micro level in the UK currently represents only 0.35% of total wind capacity[6], with potential for micro generation as high as 30-40 % of the UK’s electricity needs[7], micro wind generation capacity is expected to increase significantly. From an urban wind energy perspective, most research on the

CIBSE Sustainable Awards 2013

energy potential of such systems is somewhat biased. The performance of wind turbines tends to be assessed in ideal circumstances[8, 9]. In the urban environment, the initial cost of micro wind turbines and the locations in which units are likely to be installed (i.e. where the wind speed and direction are very dependent on the site, proximity to potential obstacles etc) have received little attention. So, the energy yield of the turbine rather than the available wind resource is studied. As a consequence, inappropriate locations for installation can never realise the energy potential. There is therefore a deficiency in our understanding of the potential energy that could be harnessed from micro wind energy systems as well as the viability of these systems to provide a costeffective power generation option[10]. Missing from most of the aforementioned research is the technology’s primary energy source, the wind. There is significant research assessing the wind energy resource in “rural” locations around the world[11-13], and in some research[14, 15] such work has been extended to apply to the potential for wind energy conversion systems. However, the available test studies that investigate the viability of micro wind generation in urban centres are more generic and broader, suggesting that the technology can work if installed correctly and in appropriate locations[8, 9]. An improved understanding of the wind resource, therefore, could potentially lead to improved choices of installation locations and more realistic productivity expectations from this form of energy system. As global populations increasingly migrate towards urban centres, small wind energy systems, either for urban dwellings or for green-field locations, must be considered as a contribution option towards renewable energy targets[16].

2. The urban wind resource and wind energy systems General adjectives that describe the wind resource include variability and unpredictability. Indeed, a full understanding of the resource is further complicated due to the fact that these characteristics are geographically and temporarily interdependent. As Burton describes[17], the variability of the wind preserves over an extensive range of scales, both in space and time. This property is enhanced in built environments where airflow is highly disturbed. Wind turbines are most efficient when airflow is strong and steady, that is high mean wind speeds with little variability in speed or direction. Urban areas disturb the airflow (reducing the mean wind speed and increasing variability) and produce sub-optimal environments for turbines. However, little is known of the actual wind resource in urban areas and a major (technical) barrier to the effective deployment of wind turbines in these areas is due to a lack of accurate methods for estimating wind speeds and energy yields at potential urban sites[18]. This is because cities are aerodynamically rough and heterogeneous and have a highly-localised and complex wind environment. Air flowing across an urban area will interact with the underlying surface and become affected by its roughness characteristics. The

result is the formation of a distinctive boundary layer that grows in depth from the upwind edge of that surface type. Where the air flows across a series of surfaces, each of a different roughness, a series of Internal Boundary Layers (IBL) form in the along-wind direction. The growth of the IBL depends on the intensity of turbulence that transmits the effects of roughness upwards and this depends on wind-speed, surface roughness and atmospheric stability[19]. The latter describes the relative tendency for an air parcel to move vertically as a result of buoyancy and is regulated by the thermal structure[19]. Whereas unstable conditions promote vertical mixing, stable conditions suppress it. However, the Irish climate is dominated by neutral conditions, so that wind-speed and surface roughness are the main factors regulating turbulence intensity[20]. Over extensive homogenous surfaces, the wind-speed (u(z)) at any height can be estimated from

u ( z) =

u* ⎛ z − zd . ln ⎜ ȡ ⎜⎝ z0

⎞ ⎟ ⎟ ⎠

(2)

where k is von Karman's constant (0.4), z is height above the ground, zo is the roughness length and zd is the displacement height or the effective zero wind speed height. The friction velocity (u*) is a measure of the shearing stress that drives the flux of momentum to the Earth’s surface. This logarithmic relationship describes wind-speed in the direction of airflow within a boundary layer where airflow has adjusted to the underlying surface. It is properly applied to extensive homogeneous surfaces (such as grass) under neutral atmospheric conditions and is valid under these circumstances to heights (z) above (zd+zo) to a limit of z* or the wake diffusion height. The zone below (zd+zo) may be described as the roughness sub-layer (RSL). Urban environments are objectively rough (Table 1) causing lower mean speeds[21-23]. However, they are also heterogeneous so that the roughness sub-layer (RSL) is very deep, extending to

Figure 2: Air-flow modelling in terms of the logarithmic model (2). This profile performs well above z*, but within the roughness sub-layer (z*<z>zHm) the associated wind is dominated by turbulent eddies making wind classification less reliable

The Small Wind Energy Estimation Tool (SWEET) – a practical application for a complicated resource

Table 1: Davenport classification of effective terrain roughness[25] Roughness Class Approx. Open

Roughness length (z0) 0.1

Rough

0.25

Very rough

0.5

Skimming

1.0

Chaotic

2.0

( Turbulence

Description of landscape Moderately open country with occasional obstacles (e.g. Isolated low buildings or trees) at relative horizontal separations of at least 20 obstacle heights Scattered obstacles (buildings) at relative distances of 8 to 12 obstacle heights for low sold objects (e.g. buildings) Area moderately covered by low buildings at relative separations of 3 to 7 obstacle heights and no high tress. Densely built-up area without much building height variation. City centres with mix of low and highrise buildings (Analysis by wind tunnel advised)

approximately twice the mean height (z*) of the roughness elements, that is the building’s height (Hzm), that is z*≥. Hzm. In the layer below z*, the logarithmic profile (2) is no longer applicable. Sunderland et al [24] showed that for Dublin, if Eqn (2) is to be used to estimate the available wind resource in the urban area, then the height of the turbine hub should be greater than the mean obstruction height in any direction by a factor of 1.5. Accounting for turbulence is more problematic, however. Turbulent flows can be described as those in which the fluid velocity varies significantly and irregularly, in both position and time[26]. While turbulently fluctuating flows impact directly on the design of wind turbines, they also influence the productivity of power within the turbines, particularly in areas of complex morphologies. Turbulence is considered further in Section 2.2.

2.1 Describing the wind resource Wind speed at a site is often described statistically. Commonly employed approaches include the Weibull and Raleigh distributions. Both have been shown to give a good fit to measured wind speed data[27]. The Weibull distribution function is described in in the following equation:

(3)

The Weibull scaling factor, c, has the same units describing wind speed, k, represents the Weibull shape parameter, ui is a particular wind speed and du is an incremental wind speed. P(u < ui < (u + du)) is the probability that the wind speed is between u and(u + du)[28]. The Rayleigh distribution is a special case of the Weibull distribution in which the shape parameter, k, has a value of 2.0[28]. From (3), the Rayleigh distribution function is :

)

(

⎛ )⎜ ⎝

⎞ ⎟ ⎠

⎡ ⎛ ⎢ ⎜⎜ ⎢⎣ ⎝

⎞ ⎟⎟ ⎠

⎤ ⎥ ⎥⎦

(4)

When wind encounters a solid unmovable object it responds by diverting in a number of ways. Firstly, the leading edge of the object will experience an increase in pressure. Secondly, the wind stream will spread out and pass around the object and recombine after the object giving the classic teardrop shape. The recombination of wind downstream from an object is never perfect and, as a result, pressure patterns in the form of vortices are produced. One concept that is often lost is that wind is the movement of a gas and, as such, it compresses and decompresses with increasing and decreasing relative pressure. Note there is a change in direction as well as a change in wind speed as it passes an object. Also, as there is now an increase in pressure at the leading edge of the object, upstream wind is diverted around this higher pressure area, thereby creating drag and slowing down the holistic wind speed. While this is a simple example, it does not cater for permeable objects such as trees or the complexity of multiple objects such as in an urban environment. Furthermore, increasing complex morphologies such as those contained in cities also lead to an increased surface roughness length characteristic, further contributing to the manifestation of more erratic wind speed signals. Turbulence Intensity (TI) is the most common metric to explain the turbulent effect on the wind. It is generally more useful to develop descriptions of turbulence in terms of statistical properties[17]. TI is defined as “the ratio of wind speed standard deviation to the mean wind speed[29], determined from the same set of measured data samples of wind speed, and taken over a specified time” and should actually be considered as the standard deviation of the wind speed σu normalised with the mean wind speed u: (5) The complex morphology experienced in an urban environment causes a modified flow and turbulence structure in the urban atmosphere in contrast to the flow over “ideal or homogenous” surfaces. With respect to the impact on the power output of wind turbines subjected to turbulence, Cochran[30] presented a description for turbulence intensity within the lower portion of atmospheric boundary layer also based on surface roughness. His conclusions were that the (kinetic) energy available at the turbine hub height can vary by as much as 20% depending on the level of TI present at a site. In[31-33] the effect turbulence intensity has on the power curve of a turbine is that high TI exaggerates the potential output power from a turbine at moderate wind speeds (cut-in), whereas low TI undermines the potential output power at rated wind speed (Figure 3). Available studies utilise measured data to provide a description of the turbulent effect and these studies are often in rural locations. However, in urban environments, unbiased high resolution wind data is difficult to acquire for wind speeds – and similarly, reliable and unbiased data for wind turbines (vis- à-vis localised building morphologies) in such environments is practically non-existent.

CIBSE Sustainable Awards 2013

(a)

Figure 3: The effect of TI on Turbine Power Curves (interpreted from [31]

3. Power predictability using high resolution (10Hz) meteorological observations

(b)

Research carried out by Sunderland et al [34] considered observations made at two urban locations in Dublin, Ireland. St Pius X National (Girls) School (SUB1), located in Terenure, Dublin 6W and Dublin City Council Buildings (URB1), in Marrowbone Lane, located in Dublin 8. URB1 is located closer to the city centre than SUB1 and is therefore more urbanised with a higher associated roughness length. At both sites, high-resolution wind speed measurements were taken. The observations were at 10Hz at an associated resolution between 0.5 and 1.0 mm/s. The analysis was based on measurements taken over a 40-day period from 4/4/2012 to 15/5/2012. Consistent with the industry standard[29], a 10-minute sampling period was employed on a moving window basis with each window consisting of 6000 samples (10 minutes at 10Hz). Two models were developed to quantify the affect of turbulence on the productivity of a wind turbine in an urban environment*. The first approach was an adaptation of a model originally derived to quantify the degradation of power performance of a wind turbine using the Gaussian probability distribution (Albers approximation) to simulate turbulence (and more specifically, turbulence intensity (TI)[35]. The second approach used the Weibull Distribution, a widely-accepted means to probabilistically describe wind speed. The advantage of the high-resolution observations in testing the models is the ability to interrogate the 6000 wind speed datums within the 10-minute interval. This facilitates inter comparison of the models against the power if it could be recorded at such high resolution and also against the standard industry approach that quantifies the turbine power output on the basis of the 10minute mean wind speed. Figure 4 illustrates a comparison of the performance of the models over the 40-day period. Each observation window considers three power measurements â&#x20AC;&#x201C; the Albers approximation Pnorm, the Weibull approximation, Pweib, and the average power over the window, Pmean, which is calculated * The wind energy system considered was a Skystream 3.7, 2.4kW turbine

Figure 4: The cumulative error for each of the calculated power models (Pmean, Pnorm and Pweib) for both sites (URB1 (a) and SUB1 (b)). Putting this in context, both the Albers or Weibull turbulence models for both sites have over 90% of its error within 50W of the Pabs at both sites

by considering the turbine characteristic with respect to the mean speed over the observation window. Pmean, is the industry norm for data logging of power output from wind turbines. Each of these calculations is benchmarked against the absolute power, Pabs, which is the average of individualised (6000) calculations. Figure 4 actually presents a cumulative sum of differences that occur throughout the full set of 40 days of data; Figure 4(a) illustrates this trend analysis for URB1; and Figure 4(b) illustrates similar for SUB1. It is clearly evident that for both sites, Pweib and Pnorm are virtually horizontal, with only a slight over-prediction derived using Pweib and under-prediction using Pnorm cumulatively derived over the 40 days of observations. This strongly implies that both models are consistent with the Pabs measurements and are accurate with respect to representing practically the effect of turbulence on the wind turbine.

The Small Wind Energy Estimation Tool (SWEET) – a practical application for a complicated resource

(a)

Figure 5: Flow chart illustrating how SWEET incorporates varying surface roughness (z0) and turbulence intensity (TI ) (this SWEET tool is available FREE at http://arrow.dit.ie/sdar/)

4. Wind energy prediction: a refined approach (SWEET) One of the aims of this paper was to describe an accessible tool (developed in MS EXCEL) that can facilitate an estimate of the wind energy based on the land surface and likely prevalence of turbulence. The Small Wind Energy Estimation Tool (SWEET) was therefore developed in this regard and is available at http://arrow.dit.ie/sdar/. More specifically, SWEET provides the user a means to approximate the annual electrical energy yield from a generic wind turbine based on the following input parameters: 1. A reference wind speed (for example one such wind resource mapping repository is available from the Sustainable Energy Authority of Ireland (SEAI) at http://maps.seai.ie/wind/#) 2. A value describing the surface roughness. The tool employs a scaling version of the log law (6)

⎛ ⎜ ⎝ ⎛ ⎜ ⎝

⎞ ⎟ ⎠ ⎞ ⎟ ⎠

(b)

(6)

where vhub is the wind speed at the hub height (zhub) and vhub and (zhub) represent the wind speed and height of the reference resource respectively 3. A hub height for the wind turbine 4. The turbine characteristic contained within the tool is generic. Users can overwrite the power curve data with alternatives if they choose 5. A turbulence intensity value based on 0.1<TI<0.7, with a TI of 0.7 Figure 5 presents a flow chart to illustrate how the tool is used. SWEET uses the wind speed reference acquired from a wind speed reference such as the one developed by SEAI and extrapolates this wind speed logarithmically to the turbine hub height as specified by the user in terms of the surface roughness, again specified by the user. The turbine power characteristic (either the generic curve contained within the tool or an alternative) is turbulence modified. The normalisation is based on a Gaussian distribution approximation developed by Sunderland et al [34]. Essentially, the turbine power at the extrapolated wind speed is modified in terms of a turbulence model which is itself based on a

Figure 6: Screen shots of SWEET in terms of varying TI

CIBSE Sustainable Awards 2013

Gaussian distribution with a mean wind speed of the extrapolated wind reference and a standard deviation acquired through (3).

flowing directly into the turbine hub and the climate is neutrally buoyant;

Looking at screen shots of the tool outputs (http://arrow. dit.ie/sdar/), Figure 6(a) illustrates the yield calculated when the reference wind speed is 5m/s and TI is 10% (for a hub height of 15m and a very rough surface (Table 1) roughness characteristic (z0=0.5)). Figure 6(b) then illustrates how the resource is affected if the turbulence intensity is increased to 60%. The yield appears to increase by 42%! This is one of the limitations of the turbulence modelling using a Gaussian distribution. The turbine power curve is skewed by the higher turbulence intensity exaggerating the potential output power from the turbine at moderate wind speed (highlighted) and, as a consequence derives a higher yield.

– The mechanical inertia of the turbine is not taken into consideration. The reaction/yawing of a turbine in fluctuating wind speed will detract from its energy producing capability;

On the other hand, if the reference wind speed is increased to 10m/s for the same input considerations (z0, zhub), the energy yield will decrease by almost 13% when the TI is 60% when compared to the yield derived with a TI of 10%.

4.1 Conclusions and further work This paper describes the complications associated with wind energy estimation, particularly in the urban environment. It is evident that the associated complexities within an urban context pose many challenges to micro wind energy installation designers and those wishing to use this form of renewable energy. In particular, what turbulence is and more importantly how it affects micro wind energy systems are significant issues. In consideration of a model that employs Gaussian statistics in[34] a refinement that can be accessed through a readily-available platform (Excel) was devised, namely SWEET. While not affording the depth of accuracy available when 10Hz data is available, this tool can serve to facilitate an estimation of wind energy yield expected at a site, as well as affording insight for engineers within the built environment to the challenges involved in optimally locating such turbines. However, there are a number of caveats associated with SWEET (as described on the first page of the tool): – Turbulence intensity is indicative for all associated site wind speeds. In this regard, the approach is predicated on the wind speed sampling being described accurately by a Gaussian probability distribution; – The wind speeds are longitudinal in nature, i.e., always

Figure 7: Wind speed trending within an observation window

– The input turbine characteristic employed is indicative of a 0% turbulent environment. A further limitation of the approach is in the consideration of TI as the metric to define turbulence, particularly in highly-turbulent environments or when the wind speed approaches 0m/s. Firstly, TI is based on Gaussian markers and the signal may or may not represent a Gaussian form. There is also potentially a problem with trending within the wind speed window being considered as turbulence when in fact it could be a gradual change in wind speed, something acknowledged by the industry standard[29], which employs the Normal Turbulence model (a precursor to the Gaussian approximated described here). Finally, and with respect to SWEET, both the surface roughness parameter and TI reference are entered separately. Mertens[36] proposes that TI can be linked to the surface roughness parameter. Such a linkage was considered and, as suggested in[34] is an area warranting further consideration. As can be demonstrated in the model there are several other factors that could be considered as a means of improving the estimation capability. Firstly, SWEET has no means to consider the most predominant wind direction for turbulence manifestation and is something that could possibly be included on a turbulence rose. Secondly, the inability of any PDF (Probability Distribution Function) markers to accurately regenerate to a time series is an issue when trying to establish the effect of inertia and its resultant time constant on the system. A novel model, the Turbulent Fourier Dimension (TDf) that measures turbulence and maintains the ability to interchange between the time domain and frequency domain is currently being investigated.

Acknowledgements The authors would like to acknowledge the support of Dublin Institute of Technology and the Dublin Energy Lab. The authors would further like to acknowledge the Dublin Urban Boundary Layer Experiment (DUBLex) collaboration involving Dublin Institute of Technology, National University of Ireland Maynooth and University College Dublin.

The Small Wind Energy Estimation Tool (SWEET) – a practical application for a complicated resource

References [1]

AEA. (2011, 12th July, 2012). The AEA Microgeneration Index [on-line]. Available: http://www.aeat.com/microgenerationindex/reports/ The%20AEA%20Microgeneration%20Index%20-%20Issue%204.pdf

[2]

Environment Community and Local Government, "Building Regulations 2011 Technical Guidance Document L: Conservation of Fuel and Energy - Dwellings," 2011.

[3]

SEAI, "Status Report on Microgeneration in Ireland," Sustainable Energy Authority of Ireland,2011.

[4]

EN 50438 Requirements for the connection of micro-generators in parallel with public low-voltage distribution networks, CENELEC, 2007.

[5]

DECC, "National Reneweable Energy Acion Plan," Department of Energy & Climate Change,2010.

[6]

Ren21, "Renewables 2012: Global Status Report," Renewable Energy Policy Network for the 21st Century, on-line2012.

[7]

Watson J., Sauter R., Bahaj B., James P., Myers L., and Wing R., "Domestic micro-generation: Economic, regulatory and policy issues for the UK,"Energy Policy, vol. 36, pp. 3095-3106, 2008.

[8]

EST, "Location, Location, Location: Domestic small-scale wind field trial report," Energy Savings Trust,2009.

[9]

Encraft, "The Warwick Urban Wind Trial Project," 2009.

[10] Arifujjaman Md, Iqbal M. T., and Quaicoe J. E., "Energy capture by a small wind-energy conversion system," Applied Energy, vol. 85, pp. 4151, 2008. [11] Islam M. R., Saidur R., and Rahim N. A., "Assessment of wind energy potentiality at Kudat and Labuan, Malaysia using Weibull distribution function," Energy, vol. 36, pp. 985-992, 2011. [12] Cabello M. and O. J. A. G., "Wind speed analysis in the province of Alicante, Spain. Potential for small-scale wind turbines," Renewable and Sustainable Energy Reviews, vol. 14, pp. 3185-3191, 2010. [13] Fyrippis I., Axaopoulos P. J., and Panayiotou G., "Wind energy potential assessment in Naxos Island, Greece," Applied Energy, vol. 87, pp. 577586, 2010. [14] Kavak A. E. and Akpinar S., "A statistical analysis of wind speed data used in installation of wind energy conversion systems," Energy Conversion and Management, vol. 46, pp. 515-532, 2005. [15] Jowder F. A. L., "Wind power analysis and site matching of wind turbine generators in Kingdom of Bahrain," Applied Energy, vol. 86, pp. 538-545, 2009. [16] Ayhan D. and Sağlam Ş., "A technical review of building-mounted wind power systems and a sample simulation model," Renewable and Sustainable Energy Reviews, vol. 16, pp. 1040-1049, 2012.

[23] Landberg L., Myllerup L., Rathmann O., P. E. L., J. B. H., B. J., and Mortensen N. G., "Wind Resource Estimation - An Overview," Wind Energy, vol. 6, pp. 261-271, 2003. [24] Sunderland K. M., Mills G., and Conlon M. F, "Estimating the wind resource in an urban area: A case study of micro-wind generation potential in Dublin, Ireland," Journal of Wind Engineering and Industrial Aerodynamics, vol. 118, pp. 44-53, 2013. [25] Oke T.R., "Initial guidance to obtain representative meteorological observations at urban sites," 2006. [26] Pope S. B., Turbulent Flows: Cambridge University Press, 2000. [27] Justus C. G., Hargraves W. R., and Yalcin A., "Nationwide Assessment of Potential Output from Wind-Powered Generators," Journal of Applied Meteorology, vol. 15, pp. 673-678, 1976/07/01 1976. [28] Seguro J. V. and Lambert T. W., "Modern estimation of the parameters of the Weibull wind speed distribution for wind energy analysis," Journal of Wind Engineering and Industrial Aerodynamics, vol. 85, pp. 75-84, 2000. [29] IEC, "International Standard 61400-2. Wind Turbines - Part 2: Design requirements for small turbines," ed, 2006. [30] Cochran B., "The Influence of Atmospheric Turbulence on the Kinetic Energy Available During Small Wind Turbine Power Performance Testing,," IEA Expert Meeting on: Power Performance of Small Wind Turbines Not Connected to the Grid,2002. [31] Langreder W., Kaiser K., Hohlen H., and Hojstrup J., "Turbulence Correction for Power Curves," presented at the EWEC, London, 2004. [32] Tindal A., Johnson C., LeBlanc M., Harman K., Rareshide E., Graves AM., and America G. H., "Site-specific adjustments to wind turbine power curves," presented at the AWEA Wind Power Conference, Houston, 2008. [33] Wagner R., Courtney S. M., Torben L. J., and Paulsen S. U., "Simulation of shear and turbulence impact on wind turbine power performance,," Riso DTU (National Laboratory for Sustainable Energy),2010. [34] Sunderland K., Woolmington T., Blackledge J., and Conlon M., "Small wind turbines in turbulent (urban) environments: A consideration of normal and Weibull distributions for power prediction," Journal of Wind Engineering and Industrial Aerodynamics, vol. 121, pp. 70-81, 2013. [35] Albers A., "Turbulence Normalisation of Wind Turbine Power Curve Measurements," Deutsche WindGuard Consulting GmbH,2009. [36] Mertens S., "Wind Energy in the Built Environment: Concentrator Effects of Buildings," Technische Universiteit Delft, PhD,2006.

[17] Burton T., Sharpe D., Jenkins N., and Bossanyi. E., Wind Energy Handbook: John Wiley & Sons, 2001. [18] Millward-Hopkins J. T., Tomlin A. S., Ma L., Ingham D. B., and Pourkashanian M., "Mapping the wind resource over UK cities," Renewable Energy, vol. 55, pp. 202-211, 2013. [19] T. R. Oke, Boundary Layer Climates, 2nd ed.: Routledge, 1988. [20] Metzger M., McKeoin B.J., and Holmes H., "The near neutral atmospheric surface layer: turbulence and non-stationarity," Philosophical Transactions of the Royal Society A: Mathematical, Physical and Engineering Sciences, vol. 365, pp. 859-876, 2007. [21] Lars Landberg, Lisbeth Myllerup, Ole Rathmann, Erik Lundtang Petersen, Bo Hoffmann Jorgensen, Jake Badger, and N. G. Mortensen, "Wind Resource Estimation - An Overview," Wind Energy, vol. 6, pp. 261-271, 2003. [22] S. L. Walker, "Building mounted wind turbines and their suitability for the urban scale—A review of methods of estimating urban wind resource," Energy and Buildings, vol. 43, pp. 1852-1862, 2011.

Areas of research

Focus on applied research The School of Electrical and Electronic Engineering (SEEE) in the Dublin Institute of Technology focuses on applied research with a strong emphasis on producing

U U U U U U U U U U U U U U

Biomedical Engineering Assistive Technology and Health Informatics Audio Engineering Wireless Comms Photonics Sensing for Structures RF Propagation Microelectronic Circuits and Systems Control Systems and Robotics Engineering Education and Teaching & Learning Pedagogy Information and Communications Security Sustainable Design Energy Management Lighting Renewable Technologies

useful and novel ideas to help Irish industry compete globally.

Each year SEEE research produces patents and technologies to licence and most recently has resulted in a spin-out company in the area of mobile communications. SEEE research is recognised for its impact and quality, which in many cases is on a par with that of the very best groups internationally. Researchers in the school have also built strong collaborations with internationally-renowned groups in Europe, India, China and elsewhere, allowing the Schoolâ&#x20AC;&#x2122;s researchers to access unique research knowledge and facilities. Contact Professor Gerald Farrell, Head of School Tel: +353 1 402 4577 Email: gerald.farrell@dit.ie

School Research Centres The Antenna & High Frequency Research Centre (AHFR) ahfr.dit.ie/ The Dublin Energy Lab (DEL) dublinenergylab.dit.ie/ dublinenergylab/ The Photonics Centre (PRC) prc.dit.ie/ The Electrical Power Research Centre (EPRC) dit.ie/eprc/ The Communications Network Research Institute (CNRI) cnri.dit.ie/

www.dit.ie/colleges/collegeofengineeringbuiltenvironment/collegeresearch/

Performance of a demand controlled mechanical extract ventilation system for dwellings

I. Pollet RENSON, BIRCHOLT ROAD, MAIDSTONE, UNITED KINGDOM GHENT UNIVERSITY, DEPARTMENT OF BIOSYSTEMS ENGINEERING, COUPURE LINKS 653, GENT, BELGIUM

J. Laverge GHENT UNIVERSITY, DEPARTMENT OF ARCHITECTURE AND URBAN PLANNING, JOZEF PLATEAUSTRAAT 22, GENT, BELGIUM

A. Vens RENSON, BIRCHOLT ROAD, MAIDSTONE, UNITED KINGDOM

F. Losfeld RENSON, BIRCHOLT ROAD, MAIDSTONE, UNITED KINGDOM

M. Reeves RENSON, BIRCHOLT ROAD, MAIDSTONE, UNITED KINGDOM

A. Janssens GHENT UNIVERSITY, DEPARTMENT OF ARCHITECTURE AND URBAN PLANNING, JOZEF PLATEAUSTRAAT 22, GENT, BELGIUM

Building Servicesnews

CIBSE Sustainable Awards 2013

Abstract The main aim of ventilation is to guarantee a good

1. Introduction

indoor air quality, related to the energy consumed for

On continental Europe, demand-controlled ventilation (DCV) is considered today as a particularly relevant alternative to other mechanical extract ventilation systems (MEV), and especially for mechanical extract ventilation systems with heat recovery (MVHR). For the moderate climate zone of Western Europe, with about 2500–3000 heating degree days, the pay-back time for investments in heat recovery ventilation is long, especially in buildings with relatively low air change rates such as dwellings.

heating and fan(s). Active or passive heat recovery systems seem to focus on the reduction of heating consumption at the expense of fan electricity consumption and maintenance. In this study, demandcontrolled mechanical extract ventilation systems of Renson (DCV1 and DCV2), based on natural supply in the habitable rooms and mechanical extraction in the wet rooms (or even the bedrooms), was analysed for one year by means of multi-zone Contam simulations on a reference detached house and compared with standard MEV and mechanical extract ventilation systems with heat recovery (MVHR). To this end, IAQ, total energy consumption, CO2 emissions and total cost of the systems are determined. The results show that DCV systems with increased supply air flow rates or direct mechanical extract from bedrooms can significantly improve IAQ, while reducing total energy consumption compared to MEV. Applying DCV reduces primary heating energy consumption and yearly fan electricity consumption at most by 65% to 50% compared to MEV. Total operational energy costs and CO2 emissions of DCV are similar when compared to

Due to its competitive price setting as well as due to reports in popular media and scientific literature about possible health risks associated with heat recovery systems[i,ii], simple central MEV dominates the residential ventilation market in this region[iii, iv]. The great variability of a dwelling occupancy in time and place enhances the potential of DCV. By applying DCV, heating energy related to ventilation is reduced by 20% to 50%, while electricity consumption is similarly reduced[v-xvii]. In the UK, due to no recognition of such an advanced system either under Part F of the Building Regulation or under Appendix Q of the Standard Assessment Procedure (SAP), with the Code for Sustainable Homes tightly tied to SAP, DCV has little or no chance on the market[i,xviii, xx]. The aim of this paper is to assess theoretically the energy saving potential of DCV and the indoor air quality (IAQ) to which the occupants of the dwelling are exposed, compared to normative ventilation systems. Two different demand-controlled mechanical extract ventilation (DCV) systems (DCV1 and DCV2) in comparison with passive stack ventilation (PSV), MEV and MVHR were investigated. In addition, an overall comparison was made between the different ventilation systems concerning annual primary energy consumption, annual CO2 exhaust, annual energy cost for ventilation heat losses and fan(s) consumption, and net present value (NPV) over 15 years.

MVHR. Total costs of DCV systems over 15 years are smaller when compared to MVHR due to lower

2. Methodology

investment and maintenance costs.

The three reference ventilation systems (PSV, MEV and MVHR) and the two DCV systems (DCV1 and DCV2) were designed on a detached dwelling, on the one hand according to the British ventilation regulation (approved document F) and on the other hand according to the Belgian ventilation regulation (NBN D 50001). The resulting design air flow rates are shown in Table 1.

Key Words: Demand, controlled, ventilation, assessment, procedure, model, simulation. AD F – Approved Document F IAQ – Indoor Air Quality DCV – Demand-Controlled Ventilation MEV – Mechanical Extract Ventilation Systems MVHR – Mechanical Ventilation with Heat Recovery PSV – Passive Stack Ventilation 40

In contrast to other countries, the design supply air flow rates according to AD F vary strongly among the different ventilation systems. Since different design flow rates have a substantial influence on the performance of a ventilation system, simulations for the UK were also carried out based on the same design supply rates of MVHR. Similar optimisation changes were carried out by Palmer et al (2009)[xviii]. In practice, it is also found that the fan(s) of a MEV or MVHR system is often set on the intermediate operating speed (or even the low

MVHR

DCV1

Living 28 Office 12 Bedroom 1 12

36 8 17

2 2 2

36 8 17

18 4 8

36 8 17

2 2 2

36 8 17

Bedroom 2 12

Bedroom 3 12

Kitchen Bathroom Toilet Utility Room Hall

12000 13889 mm² mm² 12000 13889 mm² mm² 12000 6944 mm² mm² 12000 13889 mm² mm² –

DCV2 UK

2 36 2 8 2 17 (supply) (extract) 8 8 (supply) (extract) 18 2 18 (supply) (extract) 8 8 (supply) (extract) 18 2 18 (supply) (extract) 8 8 (supply) (extract) 14 18 14

–

operating speed) instead of the high operating speed on which the design air flow rates are reached. Therefore, the MEV reference system was also simulated with half of the design extract rates as listed in Table 1. In this study, two Renson demand-controlled mechanical extract ventilation systems of (DCV1 and DCV2) based on natural supply via trickle vents in the habitable rooms and mechanical extraction in the wet rooms (such as kitchen, bathroom, sanitary accommodation (toilet) and laundry (utility)) or even the bedrooms (DCV2) were analysed (see Figure 1). Direct mechanical extraction

bathroom

kitchen

toilet

utility room

bathroom

toilet

kitchen

utility room

bedroom 2

MEV

bedroom 3

bedroom 1

PSV

extract air

Table 1 – Supply and extract design air flow rates (l/s) of the ventilation systems simulated according to the British and Belgian standard

extract air

Performance of a demand controlled mechanical extract ventilation system for dwellings

Figure 1: Configuration of DCV 1 (left side) and DCV 2 (right side)

from bedrooms can reduce the exposure to gaseous pollutants in bedrooms as studied by Laverge et al[xxi]. By means of the Belgian assessment procedure for DCV[xiv,xv], the ventilation heating energy consumption, the yearly fan electricity consumption and the realised IAQ were calculated and compared for three locations. For the UK, two locations (London and Aberdeen) with the corresponding climate were considered, Brussels was chosen as the location in Belgium. In that way, the effect of demand control in combination with the impact of the ventilation standard and the climate zone could be analysed. The effect of heat recovery used within MVHR was not considered in these simulations. Furthermore, based on the previously-calculated energy consumptions, an overall comparison was made between MEV, DCV1, DCV2 and MVHR regarding annual primary energy consumption, annual CO2 exhaust due to energy consumption, annual energy cost for ventilation heat losses and fan(s) consumption, and net present value (NPV) over 15 years. In that way also the effect of heat recovery was taken into account.

2.1 Simulation tool and parameters Both DCV systems under review and the three reference systems (PSV, MEV and MVHR) were assessed through numerical simulations with the multi-zone airflow model Contam, developed by NIST, and used within the Belgian assessment procedure for

Figure 2 Geometry of the reference building used in the equivalence procedure

CIBSE Sustainable Awards 2013

DCV. A similar model was used by Palmer et al. (2009)[xviii] to investigate the IAQ obtained with different ventilation systems, without looking to the energy performance. The geometry used in this model is based on a detached house with a ground and a first floor (Figure 2). Simulations performed on other types of dwelling (semi-detached, apartment) showed that the average effect of DCV is best approached by the detached house[xix]. The three climate zones of Brussels, London and Aberdeen with the hourly average outdoor temperature (mean outdoor temperature during heating season of London: 6.6°C; Aberdeen: 5.6°C; Brussels: 6.3°C), wind speed (mean wind speed during heating season of London: 3.3 m/s; Aberdeen: 5.2 m/s; Brussels: 5.1 m/s) and wind direction were distinguished. Climate data reveals the difference between those locations. A constant indoor air temperature of 18°C in all habitable and functional rooms was considered. A fourperson family (two parents, child and baby) with a given occupancy schedule during week and weekend is considered. Internal pollutant emission scenarios were also implemented. Windows and internal doors remained closed, while a hood was operating during cooking (56 l/s). More details about the model can be found in[xiv-xvi]. The impact of uncertainty on the previous input data can be limited by using a Monte-Carlo approach as described by Laverge et al[xiv]. Due to the significant higher simulation time of this MC approach – keeping in mind the objective of this study to show the relative potentials of DCV rather than the absolute – this approach was not used. The ventilation heating losses were determined over the heating season from 1 October to 15 April, while the yearly electricity consumption due to mechanical ventilation systems was derived from the fan power consumption of the DCV system. The fan power as a function of airflow rate for an external static pressure of 120 Pa – which is the case for well-designed ductwork – is shown in Table 2. The same power consumption was supposed for the MEV and, in the case of the MVHR, twice the electricity consumption was taken into account. Simulations were performed for five building air tightness levels (0.6; 1; 3; 6 and 12 m³/(h.m²)). The ventilation and leakage heat losses for these different air tightness levels were extrapolated to a perfect airtight building (0 m³/h.m²) to isolate the ventilation losses. The performance of the demand-controlled system with respect to IAQ was assessed on three parameters, namely exposure to carbon dioxide, exposure to odours and the humidity level: – • The average cumulative CO2 concentration (in kppm.h) for the five building airtightness levels was chosen as a marker for indoor air quality to compare ventilation systems; Table 2 – Power consumption as a function of airflow rate per fan for the MEV, MVHR and DCV systems at an external static pressure of 120 Pa Air flow rate (l/s) 0 7 14 21 28 35 42 49 56 Consumption 11.3 12.8 14.4 16 18 20.3 23.3 26.9 30.4 per fan (W)

Ex = Eref . fE

PSV

IAQx = IAQref . fIAQ IAQx

• MEV

IAQref Ex

Eref

MVHR

Figure 3: Heating energy and IAQ factor calculation for a ventilation system ‘x’ with respect to reference systems

• The monthly average relative humidity level on a thermal bridge with a temperature factor of 0.7 must be lower than 80%, in order to limit the risk on condensation and mould; • Furthermore, the exposure to odours must be lower or equal to that of the worst-performing reference system to be accepted as equivalent. If performance of the DCV system under review is equal to that of the worst-performing reference system (PSV, MEV or MVHR) for each of these parameters, it is accepted as equivalent and a heating energy factor (fE) and IAQ factor (fIAQ) (as defined below and demonstrated in Figure 3 for a system ‘x’) is determined. • The heating energy factor (fE) of a DCV is defined as the ratio of the heating season integrated ventilation heat loss of the system (Ex) and that of the reference MEV system (Eref); • The IAQ factor (fIAQ) of a DCV is defined as the ratio of the IAQ of the system (IAQx) and that of the reference MEV system (IAQref).

2.2 Overall comparison between ventilation systems For the location of London, MEV, DCV1, DCV2 and MVHR, with an average heat recovery efficiency η of 80%, were compared with respect to: • Annual primary energy consumption (kWh/year); • Annual CO2 exhaust due to energy consumption (CO2/year); • Annual energy cost for ventilation heat losses and fan(s) consumption (£/year); • Net present value over 15 years (£). Conversion factors to primary energy were 1 for natural gas and 2.5 for electricity. CO2 emission factors based on secondary energy consumption of 0.202 kg CO2/kWh gas and 0.543 kg CO2/kWh electricity were used[xxii]. The net present value or global cost CG was calculated according to EN15459: CG (τ) = CI + Σ [ Σ (Ca,i (j) x Rd(i)) – Vf,τ(j)] The following assumptions were made: • The initial investment costs CI are listed in Table 3; • The annual energy cost Ca,i (energy) and the annual maintenance cost Ca,i (maintenance) were determined for the four systems.

Performance of a demand controlled mechanical extract ventilation system for dwellings

The annual energy cost is calculated based on the Contam simulations, taking into account the assumed user energy prices for 2013 in [xxiii] of 0.035 £/kWh gas and 0.15 £/kWh electricity. The annual maintenance cost was assumed to be a percentage of the investment cost and is listed in Table 3; • The discount rate Rd was based on an inflation rate of 2% and a market interest rate of 3%; • The final value of components Vf,τ(j) was assumed to be zero; • The calculation was done over a period equal to 15 years. This results in the following formula: CG (τ) = CI + Σ (Ca,i (energy) x Rd(i)) + Σ (Ca,i (maintenance) x Rd(i)) Table 3 – Initial investment cost and annual maintenance cost for the different systems MEV Initial investment 800 cost CI (£) Annual maintenance cost Ca,i (maintenance) 16 (£) (2% CI)

DCV1 1200

DCV2 1400

MVHR 2800

30 (2.5% CI)

35 (2.5% CI)

84 (3% CI)

3. Results 3.1 Ventilation heat losses and IAQ Figure 4 illustrates the average cumulative CO2 levels against the ventilation heat losses for the three references and the two DCV systems for the three locations of Brussels, London and Aberdeen according to current Belgian and British standards. The heating energy factor (fE) and the IAQ factor (fIAQ) of both demandcontrolled ventilation systems (DCV1 and DCV2) are shown in Figure 5. The impact of the different ventilation regulations in Belgium and the UK on the IAQ and the ventilation heat losses is obviously illustrated in Figure 4. The smaller air supply rates of all UK-designed ventilation systems and the smaller extract rates of MVHR, especially explain the lower heat losses and the worse IAQ of UKdesigned systems.

Figure 4: Average cumulative CO2 concentrations (kppm.h) above 800 ppm over outdoor CO2-concentration against ventilation heat loss (MWh/year) for the reference and the DCV1 and DCV2 ventilation systems for the three locations according to current Belgian and British standards

As can be seen in Figure 4, the IAQ of PSV and MEV is always worse with respect to that of MVHR. Due to variable wind and thermal forces on the building, airflow rates are less controlled and cross ventilation can occur, especially in the case of PSV, which causes higher CO2 concentrations, especially in the bedrooms. The ventilation heat loss of reference system MEV and MVHR (η = 0%) is not equal to each other, since the ventilation heat loss is determined by extrapolating the ventilation heat loss at different building air tightness levels to an air tightness of 0 m³/h.m², and not by simulating for a building air tightness of 0 m³/h.m². In that way the impact of the building air tightness on the ventilation heat losses is, to a certain extent, taken into account. Outdoor climate differences between London and Aberdeen have a particular impact on PSV which is most affected by thermal and

Figure 5: Heating energy (fE) and IAQ factor (fIAQ) for DCV1 and DCV2 for the three locations according to current Belgian and British standards

CIBSE Sustainable Awards 2013

wind-driven forces. Due to the lower outdoor temperatures and higher wind speeds in Aberdeen, ventilation heat losses are obviously higher for Aberdeen, while IAQ is better when compared with London. When looking to DCV1 and DCV2 in Figure 4 and Figure 5 for a given location, it is clear that both DCV systems have a similar impact on the ventilation heat losses, but huge differences are observed concerning exposed IAQ. In the case of Brussels, DCV1 realises a similar IAQ compared to MEV (fIAQ = 1.07), while the CO2 concentration exceeds of DCV2 are very small. This means that DCV2 approaches closely the IAQ of MVHR. DCV1 and DCV2 reduce the ventilation heating energy by 30% and 27%, respectively, compared to MEV. When expressed compared to MVHR (η = 0%), a heating energy reduction of 48% and 46% for a DCV1 and DCV2 system, respectively, is found. Common residential MVHR realise a heat recovery efficiency of 70% to 85%, if well designed and maintained. The situation is different for the locations of London and Aberdeen. DCV1 has an unacceptable IAQ factor (6.5-7.3) when compared to the reference system MEV, while the IAQ factor of DCV2 (0.470.60) is acceptable and situated in the middle between that of MEV and MVHR. The heating energy reduction of DCV1 and DCV2 for the two locations compared to MEV is in the range of 50% to 57% and 58% to 62% when compared to MVHR (η = 0%).

Figure 7: Heating energy (fE) and IAQ (fIAQ) factor for DCV1 and DCV2 for London and Aberdeen with adapted DCV air supply rates

With the exception of DCV1, all systems comply with the humidity and odour criteria. The question arises if design airflow rates affect considerably the realised IAQ and energy savings of a DCV system compared to reference systems. Therefore, a second set of simulations was carried out with identical air supply rates (namely the air supply rates of MVHR) for all ventilation systems under consideration for the locations of London and Aberdeen, as can be seen in Figure 6. As can be deduced from Table 1, this means that design air supply rates of PSV are reduced while those of MEV, DCV1 and DCV2 are considerably increased. Furthermore, since in practice the extract fan of a standard MEV often runs on the intermediate operating speed, this configuration was also simulated. Comparing Figure 6 with Figure 4 points out that – in case of PSV with lower design air flow rates – the IAQ deteriorates and ventilation heat losses decrease as can be expected. While the higher design supply rates of MEV induce higher ventilation heat losses, they also lead to a slight decrease of the IAQ. This is due to interactions between constant mechanical and variable natural pressures that mean that room airflow rates are increasing or decreasing. Higher design supply rates on a windward façade increase the actual air flow rate, causing higher ventilation heat losses and lower CO2 levels. The opposite occurs on leeward façades. DCV can therefore offer a solution since actual airflow rate is taken into account by measuring IAQ. As illustrated in Figure 6, a MEV running on the intermediate operating speed has an IAQ level which is at least twice as bad when compared with a MEV working on the design airflow rates. This configuration also failed on the odour criterion. The ventilation heat losses are reduced by 33% to 25%. Due to higher design air flow rates, the IAQ factor of DCV1 is considerably improved from 6.5 to 1.6 and from 7.3 to 1.3, for respectively London and Aberdeen as can be seen when comparing Figure 7 with Figure 5. All factors in both figures are expressed to the references calculated according to current UK standard.

Figure 6: Average cumulative CO2 concentrations (kppm.h) above 800 ppm over outdoor CO2 concentration against ventilation heat loss (MWh/year) for the reference and the DCV1 and DCV2 ventilation systems with supply air flow rates equal to MVHR for London and Aberdeen

The IAQ levels of DCV1 are considerably lower than those of PSV and are therefore acceptable. In the case of DCV2, the IAQ becomes equal to that of MVHR systems, since an IAQ factor of zero is obtained, instead of 0.5 to 0.6. With respect to ventilation heat losses for the location of London, the energy losses of DCV1 and DCV2 with higher air supply rates

Performance of a demand controlled mechanical extract ventilation system for dwellings

increase by about half, resulting in a heating energy factor of 0.65 to 0.68 respectively, instead of 0.43 to 0.48 in the first set of simulations. This means that heating energy reduction for DCV1 and DCV2 becomes 35% and 32% compared to MEV and about 40% when compared to MVHR without heat recovery. Due to the more severe climate in Aberdeen, heating energy losses almost double due to the higher design supply rates, giving rise to a heating energy factor of nearly 0.9 for both DCV systems. Higher design airflow rates are for this case less or not justified. Applying a better zone-controlled DCV2 is advisable. For the second set of simulations, with the exception of the MEV running on the intermediate operating speed, all systems comply with the humidity and odour criteria.

Figure 9: Annual primary energy consumption (kWh/year) of MEV, DCV1, DCV2 and MVHR(η = 80%)

3.2 Fan(s) consumption

• Annual primary energy consumption (kWh/year);

Furthermore, the annual fan(s) electricity consumption of the several mechanical ventilation systems under consideration is illustrated in Figure 8. One is designed according to AD F and one with adapted air supply rates equal to those of a MVHR system, for the location of London at a building airtightness of 3 m³/h.m². The impact of the design supply rates on the fan consumption is negligible. Only in the case of DCV, the fan consumption is slightly decreased when design supply airflow rates are higher. Due to more natural ventilation by means of cross ventilation, the average extract rate is slightly decreased.

• Annual CO2 exhaust due to energy consumption (kg CO2) • Annual energy cost for ventilation heat losses and fan(s) consumption (£/year); • Net present value over 15 years (£). As illustrated in Figure 9, primary energy consumption of MVHR is about half that of MEV without demand control. MVHR has a higher primary energy consumption for operation of the fans than for compensating the ventilation heat losses. Fan consumption of MVHR is quite high due to double fans and higher air resistance due to the heat exchanger and the filters. Demand control on MEV can considerably decrease primary energy consumption, and even give rise to a primary energy consumption similar to that of MVHR. This reduction is caused by smaller ventilation heat losses in combination with smaller fan consumption. The primary energy consumption to compensate for ventilation heat losses is about three to two times higher for DCV1 and DCV2 respectively, when compared to MVHR. However, the primary fan consumption of DCV1 and DCV2 is about 30% when compared with MVHR.

Figure 8: Annual fan(s) electricity consumption for the several mechanical ventilation systems for the location of London at a building air tightness of 3 m³/h.m², according to UK standards (left) and with adapted air supply rates (right)

The annual electricity consumption of MVHR (460 kWh) is twice that of MEV (230 kWh), since it was supposed that the specific fan power of MVHR was double of MEV. In reality, due to the presence of a heat exchanger and filters, fan consumption of MVHR can significantly be higher than supposed. By means of demand control, the average extract rate of DCV1 was reduced by about 66%, resulting in an auxiliary energy reduction of about 40%. In the case of DCV2, the average airflow rate was somewhat higher, resulting in a slightly higher electricity consumption when compared with DCV1.

3.3 Overall comparison between ventilation systems For the location of London, MEV, DCV1 (with supply rates equal to MVHR), DCV2 and MVHR (average heat recovery efficiency η of 80%) were compared in Figures 9, 10, 11 and 12 with respect to:

The annual CO2 exhaust related to the energy consumption of the several ventilation systems shows a similar trend as can be seen in Figure 10. Demand control reduces strongly the CO2 exhaust of MEV to an equivalent CO2 level of that of MVHR. The annual total energy cost of the ventilation systems was compared in Figure 11. Due to high electricity prices compared with natural gas per kWh, DCV systems have similar and even lower total energy costs when compared with MVHR, for acceptable or similar levels of IAQ. The annual energy cost of DCV and MVHR ranges between £75 and £100. This cost is about 10% of the total annual energy costs of a one-family dwelling of £800 to £1000. Figure 12 clearly illustrates that the energy cost to ventilate cannot be considered separately from the total cost of a system, including investment (product and installation cost) and maintenance cost (cleaning, sensors, filters). MEV systems with or without demand control show the lowest net present value, which is about half that of MVHR systems. Saving on the investment and maintenance cost of MVHR is done in practice at the expense of IAQ and acoustic comfort.

CIBSE Sustainable Awards 2013

MVHR, without increasing supply airflow rates. Due to the automatic control of DCV systems, the guarantee on good IAQ when applying a DCV system should not be lower than using a fully-mechanical MVHR system that is manually operated.

Figure 10: Annual CO2 exhaust (kg/year) of MEV, DCV1, DCV2 and MVHR(η = 80%)

Demand control can bring a standard MEV system to a similar level as MVHR when considering IAQ, CO2 exhaust, primary energy consumption and energy costs. Besides, due to the automatic detection of the IAQ in the different rooms, the guarantee on good IAQ is higher when compared with a manually-operated mechanical system without sensors. The total cost or net present value of qualitative MEV systems with or without demand control is nearly half that of a qualitative MVHR system, due to the higher investment and maintenance cost of the latter. Further research should focus on the embedded carbon of the system and the impact of regular filter cleaning and replacement in the case of HR, optimising the DCV system with respect to design airflow rates, and control algorithms. A Monte-Carlo approach can be applied to eliminate the uncertainties on input parameters and the effect of other UK climate zones on the performance of DCV can be analysed.

Figure 11: Annual energy costs of MEV, DCV1, DCV2 and MVHR(η = 80%)

Figure 12: Net present values over 15 years of MEV, DCV1, DCV2 and MVHR(η = 80%)

4. Conclusions By means of simulations the significant effect of demand control on the performance of a MEV system was illustrated and discussed. From the simulations, it is clear that outdoor climate can be an important parameter to take into account. The less controlled the system, the higher the impact of the outdoor climate (temperature, wind speed and wind direction) and vice versa. Under more severe climate conditions such as Aberdeen, controlling the air extraction from the bedrooms is advisable as realised within DCV2. Under certain circumstances, higher design airflow rates are needed to obtain similar IAQ levels as MEV and MVHR systems, since reference supply airflow rates of MEV are small in the UK (Table 1). When extracting and controlling airflow rates from all functional rooms and also from the bedrooms, IAQ is good, while ventilation heat losses are more than halved when compared with MEV or

Performance of a demand controlled mechanical extract ventilation system for dwellings

References [i] Crump, D., Dengel, A., Swainson, M. (2009). Indoor air quality in highly energy efficient homes – a review. NHBC foundation – Zero Carbon Hub, ISBN 978-1-84806-104-0, 84 p.

[xxiii] UK Future Energy Scenarios. National grid (November 2011). http://www.nationalgrid.com/NR/rdonlyres/86C815F5-0EAD-46B5-A580A0A516562B3E/50819/10312_1_NG_Futureenergyscenarios_WEB1.pdf

[ii] Meijer, A., Verkade, A-J., Merckx, A-M., Duijm, F., Bolscher GHT. (2010). The effect of improvement measures for ventilation on the indoor environment and health complaints in Vathorst. Clima conference 2010, Antalya. [iii] Wouters, P., Carrié, R., Jardinier, M., Savin, J-L., Laverge, J., Hartmann, T., de Gids,W., Piriou, J. (2012). DCV webinar: Demand-Controlled Ventilation in the European context: approaches in 4 countries and at EU level, November 26th 2012. [iv] Durier, F (2008). Trends in the French building ventilation market and drivers for changes. AIVC ventilation information paper 19, 8 p. [v] Afshari, A. (2003). Humidity as a control parameter for ventilation, Indoor and Built Environment, 12, 215-216. [vi] Pavlovas, V. (2004). Demand controlled ventilation – a case study for existing Swedish multifamily buildings. Energy and Buildings, 36(10), 1029-1034. [vii] Jacobs, P. (2004). Demand controlled ventilation applicable for any airtightness level and occupancy?, AIVC conference 2004. [viii] Savin, J.L. (2006). Management of the time-distribution of the needs for indoor air renewal in humidity sensitive ventilation, AIVC conference 2006. [ix] Van Den Bossche, N., Janssens, A., Heijmans, N., Wouters, P. (2007). Performance evaluation of humidity controlled ventilation strategies in residential buildings. Thermal performance of the exterior envelopes of whole buildings X. Clear-water; 7 p. [x] Woloszyn, M. (2009). The effect of combining a relative-humidity-sensitive ventilation system with the moisture-buffering capacity of materials on indoor climate and energy efficiency of buildings. Building and Environment, 44(3), 515-524. [xi] Savin, J.L. (2009). “Performance” project: improvement of the ventilation and building air tightness performance in occupied dwellings in France, AIVC conference 2009. [xii] Krus, M. (2009). Calculation of the primary energy consumption of a supply and exhaust ventilation system with heat recovery in comparison to a demand-based (moisture-controlled) exhaust ventilation system, AIVC conference 2009. [xiii] Nielsen, T.R., Drivsholm, C. (2010). Energy efficient demand controlled ventilation in single family houses. Energy and Buildings, 42(11), 1995-1998. [xiv] Laverge, J, Van Den Bossche, N, Heijmans, N, Janssens, A. (2011). Energy saving potential and repercussions on indoor air quality of demand controlled residential ventilation strategies. Building and Environment, 46(7), 1497-1503. [xv] Savin, J-L., Laverge, J. (2011). Demand-controlled Ventilation: an outline of assessment methods and simulations tools. AIVC-tightvent conference 32. [xvi] Laverge, J., Pattyn, X., Janssens, A. (2013). Performance assessment of residential mechanical exhaust ventilation systems dimensioned in accordance with Belgian, British, Dutch, French and ASHRAE standards. Building and Environment, 59, 177-186. [xvii] Santos, H., Leal, V. (2012). Energy vs. ventilation rate in buildings: a comprehensive scenario-based assessment in the European context. Energy and Buildings, 54, 111-121. [xviii] Palmer, J., Orme, M., Pane, G., Ridley, I. Davies, M. Oreszczyn, T., Lowe, R. (2009). Investigation of Ventilation Effectiveness. BD2523, HMSO, 67 p. [xix] Van den Buys, D. (2012). Equivalence for demand controlled ventilation: sensitivity study of the assessment methodology. To be published. [xx] Irwin, C. (2012). How much ventilation does this room need? Cibse Journal, September, 45-48. [xxi] Laverge, J., Novoselac, A., Corsi, R., Janssens, A. (2013). Experimental assessment of exposure to gaseous pollutants from mattresses and pillows while asleep. Building and Environment, 59, 203-210. [xxii] http://www.eumayors.eu/

Building your business requires trusted partners Today, more than ever, good business is about mutuallybeneficial and well-balanced trading partnerships. Creating, sustaining and growing such partnerships is a demanding process that, in addition to the delivery of quality products and services, requires informed communication. Existing and potential clients need to know about, and fully understand, what you provide. Building Services News is the means by which to do that. We are the partner that bridges that communications gap and helps you cement the partnerships that underpin your business.

Building Services News delivers results

Building Services news Visit our website:

buildingservicesnews.com Find us on Facebook

CIBSE-ASHRAE advert:Layout 1

17/10/2013

16:40

Page 1

Moving to a New World of Building Systems Performance

CIBSE ASHRAE TECHNICAL SYMPOSIUM IN DIT KEVIN ST CAMPUS, DUBLIN Dates: 3/4 April 2014 CIBSE

AND

ASHRAE have received

the largest ever number of abstracts for the annual Technical Symposium that will be held on 3/4 April 2014 at the Dublin Institute of Technology, Kevin St. The abstracts received from around the world included a mix of academic and industry submissions, reflecting the aim of the symposium to facilitate information-sharing and networking across practitioners and those developing understanding and knowledge across the built environment. The symposium will commission around 50 authors to present, all of whose papers will be peer reviewed.

CIBSE is seeking volunteers to act as referees so members or Fellows who are willing to review papers should contact CIBSE providing a few sentences by way of overview of their practical areas of technical expertise. As well as helping to ensure that the papers presented are of a high standard, it will give referees a chance to find out about some of the developments that will be presented at the event. If you are interested in helping please email groups@cibse.org The symposium's main theme â&#x20AC;&#x153;Moving to a New World of Building Systems Performanceâ&#x20AC;? will give a platform to the latest practice and research from around the world in active and passive building systems that will shape the future for the built environment while striving to minimise resource impact. CIBSE President George Adams said: "The CIBSE ASHRAE Technical

Symposium is an exciting two-day event which tackles a fascinating range of cutting-edge subjects. It is a unique opportunity for members and industry experts to share knowledge on, and debate, important issues in the built environment such as the adaptation of cities to the impact of population growth and climate change." ASHRAE President Bill Bahnfleth said: "ASHRAE is very pleased to collaborate in the continuation of this series of international technical symposia. I personally look forward to being in Dublin to participate in this forwardlooking exchange on the future of building systems and the building industry itself. From modeling to cutting-edge systems for both new and existing buildings to workforce development, there is something for every built environment professional in this program." A joint enterprise with ASHRAE, the symposium is also supported by the Future Cities Catapult, a newlyestablished global centre of excellence on urban innovation. For more information visit www.cibse.org/symposium2014

CIBSE President George Adams

www.cibse.org/symposium2014

ASHRAE President Bill Bahnfleth

CIBSE Ireland Region …

… just a click away CIBSE Ireland’s interactive website gives a comprehensive overview of the Institution’s aims, objectives, officers and committee members, along with details of its extensive CPD programme and technical evenings. It also includes regular news updates, and reports on inter-association activity, industry awards, participation in Government consultation bodies, and other promotional activity on behalf of the building services industry.

CIBSE Ireland is the leading organisation for information, guidance and advice on all building services related matters. Membership brings many benefits, including access to the full suite of CIBSE publications available online via the knowledge portal. For more information on how to become a member, or to progress to a higher grade of membership, log on now.

w w w. c i b s e i re l a n d . o rg

THE IRISH LIGHTER/ YOUNG LIGHTER COMPETITIONS The Irish Lighter and Young Lighter Awards are annual applied research events promoted jointly by CIBSE and the School of Electrical & Electronic Engineering in DIT Kevin St. They are open to all building services professionals, with SLL and ILP members particularly

Projects must be located in Ireland, and submissions can also be made which are based on lighting research. Best abstracts are selected by a distinguished international panel of assessors and a shortlist of entrants is then invited to submit full papers. For the Irish Lighter Award, entries are encouraged from experienced lighting designers, or engineers who can present a paper about a finished project. • There may be post-occupancy evaluation evidence that is analysed critically and provides insight for the professional lighting community; •

There may be an innovative and/or sustainable design that is at the industry cutting edge;

• Or it may be something worth publishing that will be of interest, and benefit, to the professional community. The Irish Young Lighter competition began in DIT in 2003 when the first students on the programme in Electrical Services Engineering graduated. Ken Winters was the inaugural overall winner and he then went on to represent Ireland at the international Young Lighter in London in 2004, where he won the Best Presentation. Published research papers by winners of both the Irish Lighter and Young Lighter competitions may also feature in the SDAR Journal.

encouraged to participate. Who to contact michael.mcdonald@dit.ie or kevin.kelly@dit.ie

Building Servicesnews

Entry Deadline 16 December 2013

required to critically evaluate reallife data, and to examine both successes and challenges within leading-edge projects throughout Ireland or further afield.

THE SDAR AWARDS are annual applied research events promoted jointly by CIBSE and DIT. They are supported by Building Services News, and sponsored by John Sisk & Son.

Collaborations between industry and academia allow the building services profession to develop and underpin leading-edge work with evaluation, thus creating a platform for the growth of innovative technologies in the green economy.

SDAR stands for Sustainable Design & Applied Research and it applies to engineering of the built environment. The SDAR Awards are different to other competitions in that they are intended to encourage applied research, disseminate knowledge gained from this research, and raise the level and quality of innovation in projects. Entries for the SDAR Awards are

Post-occupancy evaluations and similar critical appraisal of projects facilitates the transition from ideologically-driven innovations, sometimes offering poor value, to evidence-based applied research that proves value or identifies weaknesses. These successes and failures help inform the professional community.

http://arrow.dit.ie/sdar/

Short abstracts (or ideas of about 100 words) for entry into the SDAR Awards 2014 must be submitted by Monday, 16 December 2013, by email directly to Michael McDonald and/or Kevin Kelly of DIT at michael.mcdonald@dit.ie and kevin.kelly@dit.ie From the abstracts submitted, a shortlist will be selected by peer review, and those selected will be invited to prepare final papers by Monday, 13 January 2014. The final will take place in early March 2014 in DIT, Kevin Street. Published research papers by winners of the SDAR Awards may also feature in the SDAR Journal â&#x20AC;&#x201C; http://arrow.dit.ie/sdar/ For further information contact: michael.mcdonald@dit.ie or kevin.kelly@dit.ie

Supported by :

The School of Electrical and Electronic Engineering, Dublin Institute of Technology, is the largest education provider in the electrical and electronic engineering space in Ireland in terms of programme diversity (Apprentice to PhD), staff and student numbers. Based in Dublin city centre (Kevin Street) and established since 1887, it prides itself in providing practice-based and professionally-accredited programmes. The School offers a variety of full-time and part-time programmes, including: Level 9 (Masters) U MSc – Energy Management DT711 or DT015 U ME – Sustainable Electrical Energy Systems DT704 or DT705 U MSc in Electronic and Communications Engineering DT085 or DT086

Level 7 U BEngTech in Electrical Services Engineering, Electrical in Control and Automation Systems, Electronic and Communications Engineering U BTech in Networking Technologies

Placement Opportunities

Level 8 (Hons) U BE in Electrical and Electronic Engineering DT021 U BE in Computer and Communications Engineering DT081 U BSc in Electrical Services and Energy Management DT712 or DT018

The School’s honours-degree programmes (DT021 and DT081) offers students the chance to undertake a sixmonth industry placement in their third year of study. Students are available from March – August. If you are interested in participating in the placement programme please contact: deirdre.staunton@dit.ie or michael.core@dit.ie For further information on the school contact School of Electrical and Electronic Engineering, Dublin Institute of Technology, Kevin Street, Dublin 8 Tel: + 353 1 402 4575/4550 Email: elec@dit.ie

www.dit.ie/colleges/collegeofengineeringbuiltenvironment

Power comes in many forms.

The power to control your energy costs is just one of them. At Electric Ireland, we understand that no matter what business you’re in reducing costs is essential to your success. We also know that, when it comes to payment plans, one size doesn’t ﬁt all. That’s why we offer you very competitive ﬁxed or variable rates that take account of your individual company needs. Talk to our dedicated business energy services team about reducing your energy costs and our online tools that help you manage your energy consumption. And if your business uses gas, ask about our competitive dual fuel plan too. Call the business team today on

1850 30 50 70 electricireland.ie/business

Ireland’s energy 155038_EI_SME_Bridge_Built Env Irl Mag_FA_BD.indd 1

19/06/2013 14:29