Sdar journal 2016 web file by Pat Lehane

Issue 6 December 2016

Journal

Sustainable Design & Applied Research in Engineering and the Built Environment

The SDAR Journal is a scholarly journal in sustainable design and publishes peer reviewed applied research papers. 2016 Cover SDAR Journal.indd 1

08/11/2016 14:20

www.cibseireland.org

fastrack to the top Join

CIBSE Ireland Boost your work opportunities by joining the group of relevance

You need to be in the room with the group that supports your designs, accredits your course and sets the highest standards for your industry

www.cibseireland.org New Cibse Advert 1.indd 1

10/11/2016 08:40

Contents 5 An investigation into the cost optimality of the

Passive House retroﬁt standard for Irish dwellings using Life Cycle Cost Analysis Daniel Coyle BA (Hons) BArch MSc (ERT) MRIAI School of Architecture, Dublin Institute of Technology, Bolton Street. info@danielcoylearchitects.ie

17 Hygrothermal Risk Evaluation for the Retroﬁt of a Typical Solid-wall Dwelling

Beñat Arregi BArch (UPV/EHU) Building Life Consultancy barregi@buildinglifeconsultancy.com Joseph Little MRIAI, MSc Archit. AEES, BArch (NUI) Dublin School of Architecture, DIT. joseph.little@dit.ie

29 Assessment of two methods of enhancing

Introduction This sixth edition of the SDAR Journal highlights more than ever, not just the changing face of building services, but also the strength of the collaboration between CIBSE Ireland and DIT in responding to, driving and inﬂuencing that change. The industry is awash with new thinking, innovative technology and fresh ideas but, what has been lacking prior to the SDAR Journal is the ready availability of evidence-based research data detailing how these initiatives affect, and can be applied, to deliver the optimum indoor built environment. This edition incorporates an incredible array of papers on a diverse range of topics, all of which reﬂect exhaustive research and enlightening conclusions. As such it is an invaluable tool and design aid for all sectors of the building services industry. The researchers, authors, reviewers and editorial team deserve great credit for collaborating so closely and effectively to bring this wealth of evidence-based research data to the industry at large in such an easy-to-access and user-friendly format.

thermal mass performance of concrete through the incorporation of phase-change materials Dervilla Niall, Dublin Institute of Technology and Trinity College, University of Dublin, Ireland. Roger P. West, Trinity College, University of Dublin, Ireland. Sarah McCormack, Trinity College, University of Dublin, Ireland. dervilla.niall@dit.ie

39 Irish Large Scale Solar PV Opportunities:

A viability analysis prioritising the inﬂuence of System Harmonics

Brian West Chairman, CIBSE Ireland

Keith Sunderland, Dublin Institute of Technology, Ireland. ahogan@premiumpower.ie Andrew Hogan, Premium Power, Ireland keith.sunderland@dit.ie

The SDAR Journal is one very excellent example of the strong collaborations

49 A reassessment of General Lighting Practice Based on the MRSE Concept Christopher Cuttle MA Kit.cuttle@xtra.co.nz Editor: Dr Kevin Kelly, DIT and CIBSE Contact: kevin.kelly@dit.ie Deputy Editor: Dr Keith Sunderland, DIT Contact: keith.sunderland@dit.ie Editorial Team: Kevin Kelly, Pat Lehane, Yvonne Desmond, Kevin Gaughan, Keith Sunderland. Reviewing Panel: Prof David Kennedy, Dr Alan Hore, Dr Christopher Sanders, Ciara Aherne, Dr Avril Behan, Dr Shane Colclough, Fergus Sharkey, Dr Emma Robinson, Dr Martin Barrett, Dr Keith Sunderland, Kevin Gaughan, Prof Gerald Farrell, Prof John Mardaljevic, Dr Kevin Mansﬁeld, Dr Marek Rebow. Upload papers and access articles online: http://arrow.dit.ie/sdar/

that exist between the Dublin Institute of Technology (DIT) and the sectors it serves to provide evidence-based insights of practical use to the building services industry. Building services engineering is undergoing a period of unprecedented change. It is driven by the pursuit of holistic means of achieving net-zero energy buildings, the advent of new technologies, and the increasing ubiquity of information technology in the conception, design, operation and occupancy of buildings. DIT itself is also changing fundamentally. The process leading to DIT’s designation as a Technological University will facilitate new activities in scale and reach that build on its strong tradition with broader capabilities. The occupation and further development of the new Grangegorman campus is well underway. The ﬁrst new building – the Greenway Hub – is home to the Environmental Sustainability and Health Institute (ESHI). ESHI researches all aspects of environmental impact and, importantly, one of its most constituent research groups is the Dublin Energy Lab (DEL), the leading centre for research on energy policy, systems and devices. I want to congratulate the SDAR Journal’s editorial team on producing an excellent publication that, by straddling research and practice, has become increasingly inﬂuential to the development of the future of building services engineering.

Published by: CIBSE Ireland and the College of Engineering & Built Environment, DIT Produced by: Pressline Ltd, Carraig Court, George’s Avenue, Blackrock, Co Dublin. Tel: 01 - 288 5001/2/3. email: pat@pressline.ie Printed by: Swift Print Solutions (SPS) ISSN 2009-549X © SDAR Research Journal Additional copies can be purchased for F50

SDAR Intro pages.indd 1

Professor Brian Norton President, DIT

08/11/2016 14:43

SDAR Journal 2016

Introduction The SDAR Journal is a scholarly journal in sustainable design of the built environment and publishes peer-reviewed applied research papers for a worldwide audience. The SDAR journal is provided in print form to 2,000 subscribers annually and is listed in the Directory of Open Access Journals where it had 5,000 full paper downloads last year. The SDAR journal was launched in 2011 and is a collaboration between the Dublin Institute of Technology (DIT) and the Chartered Institution of Building Services Engineers (CIBSE) Ireland. The emphasis is on moving professionals in the built environment and energy sector from ideologically based green initiatives towards evidence-based sustainable engineering and built environment solutions. There are no author fees involved and all submissions are subjected to multiple blind peer review. The journal particularly encourages younger researchers and working engineers to publish their ďŹ ndings from innovative design practice or research, where there is adequate data analysis and evidence to support their conclusions. Critical reďŹ&#x201A;ection and objective evaluation of real-world projects are at its heart. Would-be authors are also encouraged to submit papers to the annual SDAR Awards and Irish Lighter competitions, details of which are included in this journal.

Dr Kevin T. Kelly C Eng FCIBSE FSLL FIEI Head of School of Multidisciplinary Technologies Dublin Institute of Technology Past President Society of Light & Lighting Kevin.kelly@dit.ie

Editorial Board Professor Brian Norton Dublin Institute of Technology

Professor John Mardaljevic Loughborough University

Professor Andy Ford London South Bank University

Professor Michael Conlon Dublin Institute of Technology

Professor Tim Dwyer University College London

Professor David Kennedy Dublin Institute of Technology

Dr Hywel Davies CIBSE

Dr Kevin Kelly Dublin Institute of Technology, CIBSE, SLL

Mr Brian West Chairman, CIBSE Ireland Professor Gerald Farrell Dublin Institute of Technology 2

SDAR Intro pages.indd 2

08/11/2016 14:43

SDAR Journal 2016

A Reader’s Guide In this issue we have five papers – the first is a Passive House evaluation, the second examines the hygrothermal risks when retrofitting a Georgian house in Dublin, the third assesses two methods of enhancing thermal mass of concrete using phase-change materials, the fourth analyses the viability of photo voltaic systems in Ireland and finally an international researcher suggests a new methodology for interior lighting aimed at improving quality and reducing energy. The first paper examines an economic analysis of a passive house retrofit project in Galway City, Ireland, using Life Cycle Cost Analysis (LCCA). This is a deep retrofit of a semi-detached house involving significant capital investment to achieve Passivhaus standard and significant reductions in energy usage; however, the payback is long. The critical impact of discount rate, investment time frame and energy price escalation on cost optimality for deep retrofit projects is highlighted. The second paper provides evidence that current mainstream guidance for assessing moisture risk of insulation retrofits in Ireland and the UK is unsuitable for traditional solid-walled buildings. This guidance is still based on simplified hygrothermal risk assessment methods, despite the availability of more advanced numerical software for two decades and a relevant standard in place since 2007, EN 15026. A brickfaced traditional dwelling in Dublin has been selected as a case study, and four scenarios have been simulated: its original condition and three retrofit

approaches. Results indicate that the moisture content at the base of the wall increases in all retroﬁt scenarios examined and the assemblies with high vapour permeability and no membranes result in the lowest hygrothermal risk. The third paper examines the use of thermal mass of a building to store heat and/or cold to reduce energy demand from heating and/or cooling systems. The thermal storage capacity of concrete was actively enhanced by integrating phasechange materials (PCMs) which provide a high latent heat storage capacity. The panels containing PCM displayed signiﬁcantly greater thermal storage capacity, despite having reduced thermal conductivity and density, and the panel containing lightweight aggregate/PCM composite is more effective at providing additional thermal storage.

The fourth paper examines the technologistical challenges involved in designing a large scale distributed photo voltaic (PV) system. The analysis demonstrates that without proper consideration of the PV system conﬁguration and examination of how PV inverters are employed, capacity rating breaches will affect both the distribution system operator and the consumer. The ﬁnal paper is from an experienced lighting designer/researcher from New

Zealand who reassesses current lighting practice and suggests that present methods of design are over reliant on the value of illuminance on an imaginary horizontal working plane. Evidence is offered that this focus on visual performance does not result in good quality or efﬁcient lighting schemes. Instead a new and innovative metric MRSE is offered, and the recent research around it evaluated. Results to date are promising. 3

SDAR Intro pages.indd 3

08/11/2016 14:43

Institiúd Teicneolaíochta Átha Cliath Dublin Institute of Technology Dublin School of Architecture

College of Engineering & Built Environment Over the next three years the Nearly Zero Energy Buildings (NZEB) Standard will be applied to buildings built for, or rented by, public bodies; then to all new and retroﬁtted buildings. At the same time technological innovation and other environmental and building performance concerns are creating other needs and opportunities. Dublin School of Architecture has a range of postgraduate and post-apprenticeship programmes that meet these challenges. Practicing architects, technologists and engineers who wish to increase their knowledge and skills in delivering new build and retroﬁtted buildings to the impending NZEB Standard can join a range of multi-disciplinary, blended online CPD programmes. They may also wish to go further to become industry leaders and specialists in one or more areas of building performance. Applications are being accepted for September 2017. Code

Programme

NFQ level Duration

Fee

DT774

Postgraduate Certiﬁcate in Digital Analysis and Energy Retroﬁt

1 year part time

F3,000

DT774b

MSc in Energy Retroﬁt Technology

2.5 year part time

F7,500

DT775b

CPD Diploma in Thermal Bridge Assessment (NSAI recognised)

15 weeks part time

F1,500

ARCH1182 CPD Diploma in Professional Energy Skills in NZEB

10 weeks part time

F1,000

ARCH1183 CPD Certiﬁcate in Thermal Bridge and Hygrothermal Calculation for NZEB

5 weeks part time

F500

Contact: cormac.allen@dit.ie Site foreman and supervisors working for general builders and subcontractors who wish to engage better with the way information is increasingly delivered or created on site using digital and mobile technologies should attend CPD IT for Site Workers in DIT Bolton Street. Applications are being accepted for January and September 2017. Code

Programme

ARCH6001 CPD IT for Site Workers

NFQ level Duration 6

13 weeks part time

Fee F1,050

Contact: joseph.little@dit.ie For further details see:

http://www.dit.ie/architecture/programmes/

DIT School of Architecture advert 2016.indd 1

08/11/2016 14:50

An investigation into the cost optimality of the Passive +RXVH UHWURÃ&#x20AC; W VWDQGDUG IRU ,ULVK GZHOOLQJV XVLQJ Life Cycle Cost Analysis

Daniel Coyle

BA (HONS) BARCH MSC (ERT) MRIAI SCHOOL OF ARCHITECTURE, DUBLIN INSTITUTE OF TECHNOLOGY, BOLTON STREET info@danielcoylearchitects.ie

Daniel Coyle Paper.indd 5

08/11/2016 14:46

SDAR Journal 2016

Abstract The Passive House standard represents perhaps the current state-of-the-art in low-energy building design. It is often hailed by its advocates as a cost-optimal standard to be applied to both new and existing dwellings in order to achieve Ireland’s energy and CO2 reduction targets. However, meeting the rigorous standards of Passive House in existing buildings is demanding and generally requires significantlyhigher initial capital investments. This paper summarises a research study involving an investment appraisal of an individual dwelling retrofit constructed to the Passive House standard. The research aim was to determine if the Passive House standard could become a cost-optimal model for the deep-retrofit of Irish dwellings. The problem was investigated using energy analysis (DEAP v3.2) and Life Cycle Cost Analysis tools (BLCC5), applied to a real-life case study Passive House dwelling retrofit project. Total life cycle costs for the baseline (pre-retrofit) dwelling, the Passive House retrofitted dwelling, and a range of alternative retrofit scenarios were computed. An economic appraisal using Life Cycle Cost Analysis, together with sensitivity analysis, demonstrates that the deep retrofitting of an existing dwelling to the Passive House standard can become cost optimal, if longer investment periods (* 43 years), lower discount rates () 2.6%), or higher fuel inflation (* 7%) are considered. Keywords: Low-energy Retrofit, Passive House, Cost-Optimal, Life Cycle Cost Analysis.

1. Introduction The economics of energy retrofitting are based on the premise of spending-to-save – meaning additional initial capital invested today in energy-efficient refurbishment measures should be balanced by energy cost savings in the future. The aim of this research was to investigate whether it is more costeffective for an individual private home-owner in Ireland to carry out energy efficient refurbishment measures to an existing dwelling in an intensive way (i.e. to Passive House standard); or to adopt a less intensive retrofit strategy, with higher operational energy demand, but requiring lower initial capital costs. This research question was investigated by carrying out an economic evaluation, using Life Cycle Cost Analysis (LCCA), of a case study relating to an Irish dwelling retrofitted to the Passive House standard.

2. Background The existing Irish housing stock has been described as one of the worst-performing in terms of energy efficiency in Europe, with the average Irish dwelling consuming over 25,000 kWh of primary energy (Brophy et al). CO2 emissions for Irish dwellings have been stated as being 47% higher than the average dwelling in the UK and 104% higher than the EU-27 average (Ahern et al). Current and future EU energy performance policy and directives are placing a new impetus on all member states to develop cost-optimal, advanced energy-efficiency standards for both new and existing buildings, in order to deliver on energy and emissions reduction commitments. The Energy Performance of Buildings Directive (recast) outlines long-term objectives for all EU member states of decreasing the CO2 emission levels for the building sector by 80% in 2050, compared to 1990 levels (EPBD, 2010; EC, 2102). Retrofitting the existing building stock to the required standards will clearly require significant financial investments by both governments and private individuals. It is recognised within EU policy that to realise the full potential of these energy and emissions savings, the whole life cycle costs of a building over its entire life-span must be taken into account, as opposed to just focusing on initial capital investment costs (BPIE, 2013). The energy used for space heating in existing Irish dwellings on average accounts for over 67% of household delivered energy (SEAI, 2013). Given this fact, significant reductions in both energy demand and carbon emissions can be achieved with the deep-retrofit of existing dwellings in order to minimise heat losses occurring through the building fabric. The Passive House standard represents perhaps the current ultimate in such “fabric-first” low-energy building design, and is hailed by its advocates as a cost-optimal standard to be applied to both new and existing dwellings, in order to achieve the necessary energy and CO2 reductions (Passipedia, 2015). Passive House dwellings are typified by high levels of thermal insulation (very low U-values), triple-glazed high-performance windows, minimised thermal bridging (continuity of insulation layer), structural

Daniel Coyle Paper.indd 6

08/11/2016 14:46

Cost optimality of the Passive House retroﬁt standard for Irish dwellings using Life Cycle Cost Analysis

air-tightness, and the use of Mechanical Ventilation Heat Recovery systems (MVHR) to recover residual heat otherwise lost in ventilation. Meeting the Passive House (Classic) standard requires achieving an ultra-low space heating and cooling demand of no more than 15 kWh/m2 per year, or a peak heat load of 10 W/m2, as calculated using the Passive House Planning Package (PHPP). A very high level of air-tightness must also be provided in order to achieve an air-leakage rate no greater than 0.6 times the house volume per hour under a pressurisation of 50 Pascals (PHI, 2015a). A marginally-relaxed variation of the Passive House standard introduced for existing buildings – EnerPHit – stipulates a maximum space heating demand of 25 kWh/m2 per year, and an air-tightness target of 1.0 ac/h (PHI, 2015b). However, achieving the rigorous and comprehensive standards of either full Passive House or EnerPHit in existing dwellings generally requires significant intervention, and optimised fabric and component standards, and hence higher capital investment. This poses the question – do the financial savings accrued from ongoing reduced operational energy use over the whole life-span of a Passive House retrofit justify the higher initial capital investment costs? An attempt to answer this question requires economic analysis, using appropriate investment appraisal techniques. This means examining and properly quantifying all relevant capital and operational costs, occurring at different points in time, and over the whole life cycle of a building. Simple payback calculations (the amount of time it will take to recover the initial investment in energy savings) are insufficient. Simple payback ignores the future costs and benefits occurring over the complete lifetime of a building, residual values, fuel escalation, as well as the time value of money (the impact of inflation and interest rates). Life Cycle Cost Analysis is a technique that can be used to properly evaluate the total economic performance of buildings, or energy-efficiency measures over their entire life cycle (SCSI, 2012; WBDG, 2014).

3. Literature review There is debate as to whether it is more cost-effective to refurbish existing dwellings in an intensive way in order to minimise operational energy use, or whether it is better to adopt a less intensive retrofit strategy with lower initial capital costs (Versele et al). Previous studies have used LCCA to examine the total life cycle costs of different energy-retrofit standards, in order to ask the question — is the retrofit standard with the lowest operational energy costs the most cost-optimal standard? Neroutsou (2014) used LCCA to determine the most cost-effective way to refurbish the thermal envelope of a case study end-of-terrace Victorian house in London by comparing the life cycle costs of the original pre-refurbishment building, the actual as-built “regulationscompliant” retrofit standard, and a higher Passive House (EnerPHit) standard. Total life cycle costs of the Passive House retrofit were shown to be 30% higher than the regulations-compliant standard. Neroutsou concluded, however, that Passive House could become the

economically-optimal retrofit option, but only with rising energy prices, lower discount rates (< 3.5%), and longer investment lifespans (more than 33 years). An earlier Belgian study (Versele, Vanmaele, Breesch, Kein & Wauman, 2009) conducted a similar cost benefit analysis of energy retrofitting a 1950s singe-family dwelling. Four different energy performance levels for retrofitting the dwelling were considered, including Passive House. Energy costs were calculated using both PHPP and the Flemish national energy-rating tool, EPB. The study found a 92% reduction in total end-use energy could be achieved with the Passive House standard, compared with 81% from a less intensive “low-energy” standard. The cost-optimal standard varied according to the predicted rate of fuel inflation, and the investment timescale. With a low fuel inflation forecast (2%), the Passive House retrofit failed to pay for itself, even after 40 years. Passive House was shown to be cost optimal only with a (perhaps improbable) 10% energy price increase every year, and over a 30-year investment horizon. This study correlates with the findings of Audenart, De Cleyn and Vankerckhove (2008). As in Neroutsou (2014), these studies all highlight the need for treating the conclusions of LCCA with care – the calculations are based on multiple assumptions of retrofit construction costs, estimated energy savings, variable interest rates, inflation and energy price escalation which are all difficult to predict with certainty. Famuyibo (2012) applied a similar LCCA methodology, but on a larger scale in order to provide more generalised findings and policy guidance on the economic viability of applying the Passive House standard to retrofitting the entire Irish housing stock. Famuyibo used statistical sampling, stock modelling methods, and the development of a range of representative dwelling “archetypes”. This was then combined with LCA tools to try to determine the extent of national reductions in energy, life cycle costs and carbon emissions that could be achieved in retrofitting the Irish housing stock to differing standards, meeting (then-current) Building Regulation standards, as well as to a more ambitious Passive House standard. This study concluded that retrofitting the building stock to Passive House standard could reduce national life cycle primary energy-related emissions (from dwellings) by over 84%, but that both retrofitting to Current Regulations and to a higher Passive House standard have significantly higher life cycle costs than a “do-nothing” base-case scenario. These findings would seem to be at variance with Neroutsou (2014) and Versele et al (2009).

4. Research methodology 4.1 Life Cycle Cost Analysis – key concepts and standards Life Cycle Cost Analysis (LCCA) is a technique for evaluating the total economic performance of a building asset or element over its projected lifespan, or deﬁned period of analysis. It can be described as the overall cost of constructing, operating, maintaining, repairing, renewing and disposing of an asset over its entire service life (ISO 2008a). LCCA enables comparative ﬁnancial appraisals to be made, of two or more project alternatives, in order to select the one that has the lowest life cycle costs and hence is the most cost-effective over the anticipated lifespan (SCSI, 2012: WBDG, 2014).

Daniel Coyle Paper.indd 7

08/11/2016 14:46

SDAR Journal 2016

In the context of building design and retrofitting, LCCA is a powerful economic analysis tool that can be used by architects, engineers, surveyors and other construction professionals to better inform energy-related investment decisions. LCCA allows the assessment of two key investment decisions: (1) are the increased initial investment costs incurred today justified by lower operating costs in the future? and, (2) out of two or more potential investment alternatives, which is the most economical in the long run? The alternative with the lowest overall life cycle costs will be the most cost-effective choice, assuming that it satisfies all other relevant performance requirements (Fuller & Petersen, 1995). The methodology of this study is as per the international standard (ISO 15686:Part 5), and the draft EU CEN methodology: “Cost optimal building performance requirements” (ISO 2008a, ECEEE 2011). 4.2 Life Cycle Cost formula Life Cycle Costs (LCC) are in essence the sum of all capital and operational costs, occurring at various times over the life of a building or asset. The basic formula for the summation of all life cycle costs is as follows: LCC = I + OM&R + Repl - Res + E Where LCC – Total life cycle costs; I – Initial capital investment (construction) costs; OM&R – Present-value operating, maintenance + repair costs; Repl – Present-value capital replacement costs; Res – Present-value residual value, less disposal costs; E – Present-value energy costs. 4.2.1 Initial Capital Costs Initial investment costs include all direct and indirect project and construction costs associated with achieving the energy retrofit performance standard. The study involved assessing all relevant retrofit and refurbishment costs and then separating costs into “energy-efficiency costs” (retrofit or renewal works attributable to improving energy performance), and “incidental refurbishment costs” (general refurbishment, upgrade, or reconfiguration works required to the dwelling independent of any energy performance improvements). 4.2.2 Maintenance, repair and replacement costs Maintenance, repair and replacement costs are an integral part of overall life cycle costs (ISO 2008a). Annually recurring maintenance and repair costs for a dwelling will typically include boiler or heating system servicing, changing of MVHR filters, cleaning of ductwork and maintenance of air-tight seals to windows. Depending on the chosen study period and the expected life-span of the dwelling, LCCA calculations are generally required to include any future replacement costs for building elements, equipment, and systems. This requires an estimation of the service life of such components in order to anticipate maintenance and replacement cycles. ISO 15686 gives detailed guidance on service-life planning and

estimation of life expectancy for building materials and components (ISO, 2008b). Replacement costs are assumed to be in line with current capital costs, (with costs escalated to their future value). 4.2.3 Operational energy costs (DEAP) Annual operational (fuel) energy costs for all project alternatives were calculated using the DEAP (Dwelling Energy Assessment Procedure) energy analysis software. Although the case study retroﬁt dwelling was designed to meet the Passive House performance criteria using the Passive House Planning Package software (PHPP), DEAP was adopted to estimate the operational energy demand for the various alternatives. DEAP is currently the only recognised energy performance calculation tool that can be used to provide an energy performance rating and demonstrate compliance with Part L (Conservation of Fuel & Energy) of the Irish Building Regulations, in accordance with the EU Performance of Building’s Directive (EPBD Recast Directive 2010/31/EU Article 3). For the retroﬁtted casestudy dwelling a high correlation was observed between the (DEAP) predicted operational energy use, and the actual (post-occupancy) monitored energy use (Coyle, 2015). Operational fuel costs for the LCCA analysis were then obtained by multiplying the calculated annual Delivered Energy (kWh by fuel type) given in the DEAP results page, by the relevant fuel price kWh unit costs (including VAT). These unit costs were based on the current SEAI average national fuel price database (Table 1.) (SEAI, 2015). Table 1. Average current domestic ffuel costs – 1/1/2015 (Source: SEAI 2015b). Natural gas unit price €/kWh

Oil unit unit price €/kWh

Electricity unit price €/kWh

Solid fuel (coal/peat) €/kWh

0.0681

0.0755

0.2107

0.0687

4.3 Present value analysis – calculating NPV of retrofit alternatives Fundamental to LCCA is the concept of Net Present Values (all future costs converted to their present value at the start of the project, taking into account the effects of interest rates and inflation). LCCA involves looking at cash flows and costs occurring at different time periods of the life cycle of a building. In order to be able to add and compare these costs, LCCA calculations must convert all amounts to present values (the value of anticipated future-occurring costs in “today’s money”), by applying a discount rate that reflects the “opportunity cost of money over time”. For all future-occurring costs the LCCA methodology first escalates the base year costs to their anticipated future time of occurrence, based on an escalation or inflation rate, and then discounts all costs to give Net Present Value costs (SCSI, 2012). The Net Present Value (NPV) of a particular investment scenario is thus calculated using a formula combining the escalation rate (inflation), discount rate (interest), and the study period (investment period):

Daniel Coyle Paper.indd 8

08/11/2016 14:46

Cost optimality of the Passive House retroﬁt standard for Irish dwellings using Life Cycle Cost Analysis

NPV =

(Central Bank of Ireland, 2015). For general inﬂation, the historical annual inﬂation rate for Ireland, averaged over the last 20 years, of approximately 2% was used (CSO, 2015).

Σ (l + r) t=1

Where NPV – Net present value; Ct

– is the cost in year t;

– is the expected real discount rate per annum;

– is the no. of years at the occurrence of the costs;

– is the period of analysis (investment term).

4.4 Software tools to calculate NPV Using the above basic mathematical formula a simple LCCA calculation tool can be developed using an Excel spreadsheet. Alternatively, a range of LCCA software programmes are available. One such programme is the BLCC5 software (Building Life Cycle Cost Program, version 5), developed by the US National Institute of Standards & Technology (NIST), and provided freely by the US Department of Energy. The BLCC5 software requires user input of all life cycle cost data (initial capital investment costs and operational costs) as well as deﬁning the economic boundary conditions (discount rate, escalation rate, investment period, service life and residual value factor). The software will then compute (in present-value currency) total life cycle costs for each project alternative, based on the entered cost data and economic assumptions.

With respect to the energy price escalation rate, a somewhat conservative rate of 4% was initially selected for the calculations [the actual annual escalation rate for household heating oil, for example, has been shown to average at around 6% for the period 2005-2012] (SEAI, 2013). The calculations were then repeated with a range of both higher and lower fuel escalation rates. A 50-year design life for the retrofit measures was deemed as a reasonable assessment of the minimum design life of the installed energy retrofit measures. The 30-year study period was based on an assumed maximum investment term for a fixed rate residential mortgage. Residual values (40%) were then calculated using a straight-line depreciation method in accordance with both the NIST and EU cost-optimal methodology, (WBDG, 2014; EC, 2012).

5. Case study dwelling The subject of this LCCA study is a Passive House deep-retrofit of a domestic building located in Galway City, Ireland. Designed by Simon McGuinness Architect, and completed in April 2014, the house is one of only three (at the time of writing) certified Passive House retrofit projects in Ireland (PHI, 2015c). Passive House calculation, design and construction standards were adopted to produce a retrofitted dwelling with a predicted 90% reduction in operational fuel costs, primary energy demand and CO2 emissions.

4.5 Financial assumptions used for LCCA For the initial LCCA calculations the study adopted the following key ﬁnancial assumptions: Discount Rate:

General Price Inﬂation Rate:

Energy Price Escalation Rate:

Study Period (Investment Term):

30 Years

Lifespan (of retroﬁt measures):

50 Years

Residual value

40%

A calculation of the Net Present Value (NPV) for project alternatives was then performed using the above assumptions. Sensitivity analysis was also used to assess input data uncertainty and the effect of changing the key assumptions and economic parameters underpinning the calculations. The initial ﬁnancial assumptions used were in line with ISO 15686, as well as the LCCA methodology described in the EU comparative methodology framework, the Cost-optimal Regulations [Commission Regulation (EU) 244/2012], and expanded upon in the associated Cost-optimal Guidelines [Guidelines accompanying (EU) 244/2012] (EC, 2012). A discount rate of 4% was found to be an appropriate initial assumption based on Irish Central Bank historical data for average (real) interest rates for household mortgages over the last 15 years

Figure 1. Case-study building — the existing dwelling prior to, and after PH retroﬁtting works (McGuinness, 2014).

Daniel Coyle Paper.indd 9

08/11/2016 14:46

SDAR Journal 2016

5.1 Existing construction and energy performance The original (1960s) dwelling was constructed of 300 mm thick externally-rendered and internally-plastered cavity walls (uninsulated), and a timber-trussed roof with concrete tiles. The house had an (uninsulated) solid concrete floor and timber-joisted intermediate floor, with plasterboard ceilings. Windows and doors were single-glazed and aluminium-framed. The dwelling had an outdated and inefficient heating system resulting in a very poor energy performance. The calculated Building Energy Rating (BER) for the existing original building was an F rating, with a Primary Energy Use of 388 kWh/m2/yr (Table 2). 5.2 Passive House retrofit measures The retrofit design strategy follows the Passive House design principals of a super-insulated thermal envelope (insulation continuity to avoid thermal bridging), triple-glazed Passive House certified windows, and an exceptionally-high level of structural airtightness combined with an efficient whole house mechanical ventilation system with heat recovery. The retrofit fabric and systems upgrades resulted in a retrofitted dwelling with an A2 BER rating, with a calculated total primary energy demand of 43 kWh/m2/yr. (Table 2). Table 2. Key Energy Performance Characteristics

Base (Existing)

B3 ‘Shallow Retroﬁt’

Passive House

Walls

1.78

0.21

0.12

Roof

2.30

0.16

0.11

Floor

0.84

0.17

Windows

5.80

1.6

0.96

Doors

3.0

2.0

0.90

0.15

0.30

0.08

Fabric air-tightness m3/m2.hr @ 50 Pa

0.4

Ventilation system

natural

MVHR

Heating system

oil boiler

condensing boiler

Air-Water HP

Heating system efﬁciency %

292

Heating Energy Demand (kWh/m2.yr)

270

DHW Energy Demand (kWh/m2.yr)

Primary Energy (kWh/m2.yr)

388

136

U-Values (W/m2K)

Thermal Bridging factor) (y-value, W/m2K

BER Rating

Table 3. Total initial capital investment costs (four alternatives) 1. Base ‘Do Nothing’

2. Systems Upgrade

F12,500

Capital Costs

3. B3 ‘Shallow 4. Passive Retroﬁt’ House F57,441

F110,510

Table 4. Energy demand (Delivered Energy) –- kWh/yr 1. Base ‘Do Nothing’

2. Systems 3. B3 ‘Shallow 4. Passive Upgrade Retroﬁt’ House

Heating – primary

31,768

25,053

11,170

Heating – secondary

8,420

8,573

1,622

–

DHW – primary

5,354

4,115

2,450

1,173

DHW – secondary

1,471

–

Auxiliary electrical

230

335

671

Electrical lighting

1,326

626

634

Total

48,568

38,598

16,211

2,478

BER

5.3 Capital investment costs of Passive House retrofit Initial capital investment costs for the case study retrofit project were compiled and assessed in accordance with the methodology described in Section 4.2.1. The total initial capital costs were calculated in the amount of F169,580, including VAT, professional fees and ancillary costs. Separating out the costs of the Passive House (energy-saving) measures from the general refurbishment and alteration works gives costs in the order of F110,510 (F778 per m2), representing 65% of the total project costs (Table 3). 5.4 Retrofit alternatives In order to assess the Passive House life cycle costs in comparison with other less intensive (and less costly) interventions, two alternative notional retrofit scenarios were additionally examined: (1) The existing pre-retrofit dwelling with only systems upgrades (space heating and DHW) – estimated total cost F12,500, and (2) a “shallow retrofit” involving systems upgrades as well as more conservative fabric upgrades (new double-glazed windows, external wall and roof insulation, no floor replacement or insulation), and the provision of a solar hot water system (roof mounted solar panel). This alternative is calculated to have a B3 BER rating (136 kWh/m2/yr) with initial capital costs of F57,441 (F410 per m2) – approximately half the cost of the actual realised Passive House retrofit (Tables 2,3,4).

Daniel Coyle Paper.indd 10

08/11/2016 14:46

Cost optimality of the Passive House retroﬁt standard for Irish dwellings using Life Cycle Cost Analysis

6. Results and analysis 6.1 Operation energy, fuel costs and CO2 Delivered energy, CO2 emissions and operational energy costs for each of the four retrofit scenarios were calculated and compared. The results indicate an estimated 95% reduction in total delivered energy and a 90% reduction in both CO2 emissions and operational energy costs achieved in the Passive House retrofit over the original base-line (pre-retrofit) dwelling. 6.2 Total life cycle costs Total life cycle cost calculations were carried out for the Passive House retrofit as well as the three other retrofit scenarios. The LCCA computes total (present value) life cycle costs for the Passive House retrofit to be F112,924. This includes an NPV deduction of F24,689 in respect of the remaining residual value for the retrofit works. A comparative analysis between the Passive House and the original “do-nothing” base case dwelling shows that the Passive House measures are cost-effective, with predicted Net Savings (NS) in the amount of F34,626, a Savings-to-Investment Ratio (SIR) of 1.4, and an Adjusted Internal Rate of Return (AIRR) of 5.18%. Simple Payback occurs in year 18, and Discounted Payback after 28 years.

Table 5. Results of LCCA calculations for project alternatives Initial Capital Costs 1. Base – ‘do nothing’

Total LCC (PV)

Net Payback Savings (PV) Period (Discounted)

F147,550

–

2. Upgrade Systems

F12,500

F131,210

F16,341

15 yrs

3. ‘Shallow Retroﬁt’ B3

F57,441

F101,241

F46,309

19 yrs

4. Passive House

F110,510

F112,924

F34,626

28 yrs

From the comparative LCCA results (Table 5), it is evident that all of the retrofit measures have lower total life cycle costs than the “do-nothing” base dwelling, meaning they are all cost-effective, or “profitable” over the 30-year study period. On a total Life Cycle Cost basis, doing nothing is actually the most expensive option. On a purely financial basis, the LCCA suggests that the B3 “shallow retrofit” scenario is the most cost-optimal of all the alternatives considered. The LCCA calculates it to have the lowest overall life cycle costs, generating the highest net savings (F46,309). This is followed in second place by the Passive House retrofit with net savings of F34,626. The fact that the retrofit alternative involving only an upgrade of the heating system produces the lowest net savings (F16,341), despite having much lower initial capital costs and the fastest payback period (15 years), illustrates the point that payback is a poor indicator of overall cost-effectiveness, and moreover the principle in deep-retrofit economics of “spending more to save more”.

6.3 Sensitivity analysis It is apparent that LCCA is affected by a number of unpredictable economic variables fluctuating over time and hence contains an inherent degree of uncertainty (ECEEE, 2011, BPIE, 2013). Changing any one of the key assumptions or parameters in a LCCA calculation can impact dramatically on the results of any investment appraisal. LCCA therefore must also involve a series of “sensitivity analyses” in order to assess the impact of changing individually, and in combinations, all of the key economic variables such as: • Discount rate (real rate of annual interest); • Fuel inflation (escalation rate); • Investment time span (study period); • Residual values; • Variations in actual capital construction costs; • Fluctuations in actual operational energy savings. The discount rate selected is perhaps the most critical factor in LCCA calculations, and hence the cost-effectiveness of the energy retrofitting measures assessed. Low discount rates produce higher net savings, encouraging higher initial investment costs, whereas an increasing discount rate leads to decreasing present-value future savings. With a discount rate at or below 2.7%, the Passive House retrofit becomes more cost-effective (greater total net savings) than the cheaper B3 “shallow retrofit” alternative. The net savings (profits) generated by the Passive House retrofit increase to over F200,000 with a 0% discount rate, while at a discount rate above 5.6% the Passive House retrofit measures become no longer cost-effective (negative Net Present Values), (Figure 3 – top). An increasing fuel escalation rate on the other hand leads to increasing net savings from the Passive House retrofit measures. Net savings increase exponentially with increasing fuel inflation. The initial LCC calculation used a fairly conservative 4% fuel inflation rate. Although perhaps an unlikely long-term scenario, with static or falling fuel prices () 2% inflation rate), the Passive House retrofit becomes no longer economic (Figure 3 — middle). At a fuel escalation rate of around 7%, the Passive House retrofit overtakes the cheaper B3 “shallow-retrofit” alternative in terms of cost-effectiveness. Assuming a future fuel inflation rate of 10% (unlikely perhaps but possible), the profits generated by the Passive House retrofit increase nearly eight-fold to over F250,000. The longer the investment period considered, the greater the net savings generated by energy retrofitting. With a study period less than 19 years, the Passive House becomes no longer economic – operational energy savings accrued are not enough to offset the initial higher capital investment. With with a study period of over 43 years the Passive House retrofit overtakes the cheaper B3 “shallowretrofit” alternative (Figure 3 – bottom). Assuming a 100-year investment period, the net savings (profits) generated by the investment in the Passive House retrofit increase to over F300,000.

Daniel Coyle Paper.indd 11

08/11/2016 14:46

SDAR Journal 2016

7. Conclusions The primary aim of this research was to conduct an economic appraisal of the Passive House retrofit standard using Life Cycle Cost Analysis, in order to determine if Passive House could become a costoptimal standard for the deep-retrofit of Irish dwellings. The case study project analysed in this study demonstrates how a state-of-the-art, deep-retrofit of an existing dwelling can achieve advanced levels of energy performance. Energy analysis of the case study dwelling showed that reductions of over 90% in energy and CO2 emissions can be delivered in a typical “pre-regulations” Irish dwelling by deep retrofitting to the Passive House standard. Applied on a much wider scale, this offers the potential to realistically meet, and even exceed, the building-related emissions reduction targets Ireland has committed itself to delivering by 2050. The economic appraisal carried out using Life Cycle Cost Analysis suggests that the deep retrofitting of existing Irish dwellings to the Passive House standard can be cost-effective for a private homeowner, with the right combination of interest rates () 4%), fuel inflation (* 4%), long-term investment periods (* 30 years), and the inclusion of residual values. With these initial economic parameters, the LCCA calculation showed the Passive House was a cost-effective, and even profitable, investment option, generating a positive investment return over the 30-year investment time period. That said, from a purely private, micro-economic perspective, a less intensive “shallow retrofit” is likely to be more profitable, generating greater net savings over the assumed investment term. However, with lower interest rates, longer investment timescales or higher fuel inflation, Passive House can become the cost-optimal standard. The study further demonstrated that increasing the lifespan of the investment (>43 years), reducing interest rates (<2.6%), or assuming a higher rate of fuel price escalation (>7%), all increase the cost-effectiveness of the Passive House and can justify (economically) the higher capital investment.

Figure 3. Sensitivity analysis: effect of varying discount rate, fuel inﬂation rate and investment period on NPV (cost savings).

This research study was limited in scope to an analysis of the life cycle costs for an individual private house owner. Monetarisation of wider societal or environmental costs and benefits was therefore deliberately avoided. The societal perspective (such as the environmental and economic cost of greenhouse gas emissions) was not considered. Furthermore, co-benefits such as improved indoor air quality, longevity of building construction achieved by elimination of interstitial condensation risks and potential mold growth, and also resulting improvements in user’s comfort, health and amenity were excluded (even though it is recognised that there are likely to be consequential economic benefits as a result of these). This research also focused on an economic assessment of a specific dwelling retrofit. Although the limitations of a study based on an individual case study need to be recognised, the methodology and approach taken by this research could be applied on a broader scale to investigate the life cycle cost impacts of applying the Passive House retrofit standard more widely to the existing Irish housing stock.

Daniel Coyle Paper.indd 12

08/11/2016 14:46

Cost optimality of the Passive House retroﬁt standard for Irish dwellings using Life Cycle Cost Analysis

References [1] Ahern, C, O’Flaherty, M, and Griffiths, P (2013) State of the Irish Housing Stock - Modelling the heat losses of Ireland’s existing detached rural housing stock & estimating the benefit of thermal retrofit measures on this stock, Energy Policy, 55, 149-151. Available at: http://dx.doi.org/10.1016/j.enpol.2012.11.039 [Accessed 8th December 2014] Audenaert, A, De Cleyn, S.H, and Vankerckhove, B (2008) Economic analysis of passive houses and low-energy houses compared with standard houses. Energy Policy, 36, 47-55. Available at: http://dx.doi.org/10.1016/j.enpol.2007.09.022 [Accessed 6th June 2014] BPIE (Buildings Performance Institute Europe) (2011) Europe’s Buildings Under the Microscope: A country by country review of the energy performance of buildings. Buildings Performance Institute Europe. Available at: http://bpie.eu/uploads/lib/ document/attachment/20/HR_EU_B_under_microscope _study.pdf [Accessed 19th October 2014]

McGuinness, S (2014) Airtightness: the sleeping giant of energy efficiency, Passive House + Magazine (Irish Edition) Issue 7, 76-80. Neroutsou, T (2014) Lifecycle Costing of Low Energy Housing Refurbishment: A case study of a 7 year retrofit in Chester Road, London. Paper given at the 8th Windsor Conference: Counting the Cost of Comfort in a changing world, 10-13 April 2014. Windsor, UK. Available at: http://nceub.org.uk/W2014/webpage/pdfs/ extra_papers/W14013_Neroutsou.pdf Passipedia (2015) Economy and financing of efficiency: new buildings, renovation and step by step retrofit, Available at: http://www.passipedia.org/ basics/affordability/investing_in_energy_efficiency/economy_and_financing_of_ efficiency_new_buildings_renovation_and_step_by_step_retrofit [Accessed 12th April 2015] Passivhaus Institute (2015a) Passive House requirements, Available at: http:// passiv.de/en/02_informations/02_passive-house-requirements/02_passive-houserequirements.htm [Accessed 12th April 2015]

BPIE (Buildings Performance Institute Europe) (2013) Implementing the costoptimal methodology in EU countries – Lessons learned from three case studies. Buildings Performance Institute Europe. Available at: http://bpie.eu/documents/ BPIE/publications/cost_optimal_methodology/BPIE_Implementing_Cost_Optimality. pdf [Accessed 5th December 2014]

Passivhaus Institute (2015b) Passive House EnerPHit standard, Available at: http:// passiv.de/en/03_certiﬁcation/02_certiﬁcation_buildings/04_enerphit/04_enerphit. htm [Accessed 12th April 2015]

Brophy, V, Clinch, J.P, Convery, F.J, Healy, J.D, King, C, Lewis, J.O (1999). Homes for the 21st Century: the Costs and Beneﬁts of Comfortable Housing in Ireland, report prepared for Energy Action Ltd, by the Energy Research Group and Environmental Institute, University College Dublin, Energy Action Ltd., Dublin.

Pountney, C, Ross, D, and Armstrong, S (2014) A Cost-Optimal Assessment of Buildings in Ireland Using Directive 2010/31/EU of the Energy Performance of Buildings Recast, SDAR Journal of Sustainable Design & Applied Research: Vol.2: Issue 1, Article 5.

Central Bank of Ireland, (2015), Table B.2.1. Retail Interest Rates and Volumes Loans and Deposits, New Business Available at http://www.centralbank.ie/polstats/ stats/cmab/Pages/Retail%20Interest%20Rate%20Statistics.aspx [Accessed 16th April 2015]

SCSI (Society of Chartered Surveyors of Ireland) (2012), Guide to Life Cycle Costing, Available at: http://arrow.dit.ie/cgi/viewcontent. cgi?article=1002&context=beschrecrep [Accessed 28th November 2014]

CSO (Central Statistics Office) (2015), Consumer Price Index - Annual Average, % Change 1977-2013, CSO, Cork. Available at: http://www.cso.ie/en/statistics/ prices/consumerpriceindex/ [Accessed 26th March 2015] Coyle, D (2015) A Life Cycle Cost Analysis of an Irish Dwelling Retrofitted to Passive House Standard: Can Passive House Become a Cost-Optimal Low-Energy Retrofit Standard? MSc dissertation, Dublin Institute of Technology, Dublin. Ireland. ECEEE (European Council for an Energy Efficient Economy) (2011) Cost optimal building performance requirements: Calculation methodology for reporting on national energy performance requirements on the basis of cost optimality within the framework of the EPBD. European Council for an Energy Efficient Economy, Stockholm. EPBD (2010) Directive 2010/31/EU of the European Parliament and of the Council of 19 May 2010 on the energy performance of buildings (recast). Available at: http://eur-lex.europa.eu/LexUriServ/LexUriServ. do?uri=OJ:L:2010:153:0013:0035:EN:PDF [Accessed 3rd September 2014]

Passivhaus Institute (2015c) Passive House Database, Available at: http://www. passivhausprojekte.de/index.php?lang=en# [Accessed 12th April 2015]

SEAI (Sustainable Energy Authority of Ireland) (2013). Energy in the residential sector, 2013 Report Dublin: Energy Policy Statistical Support Unit, Sustainable EnergyAuthority of Ireland. Available at: http://www.seai.ie/Publications/ Statistics_Publications/Energy-in-the-Residential-Sector/Energy-in-the-ResidentialSector-2013.pdf [Accessed 12th November 2014] SEAI (Sustainable Energy Ireland) (2015) Domestic Fuels, Comparison of Energy Costs Available at: http://www.seai.ie/Publications/Statistics_Publications/Fuel_ Cost_Comparison [Accessed 11th March 2015] Versele, A, Vanmaele, B, Breesch, H, Kein, R, and Wauman, B (2009) Total Cost Analysis for Passive Houses. Catholic University College Ghent, Dept. of Industrial Engineering, Ghent, Belgium. WBDG (Whole Building Design Guide) (2014) Resource Page: Life cycle Cost Analysis (LCCA), http://www.wbdg.org/resources/lcca.php?r=blcc [Accessed 14th December 2014]

EC (European Commission) (2012) Guidelines accompanying Commission Regulation (EU) No. 244/2012 of 16 January 2012 on the Energy Performance of Buildings by Establishing a Comparative Methodology Framework for Calculating Cost-Optimal Levels of Minimum Energy Performance Requirements for Buildings and Building Elements. Famuyibo, A (2012) Reducing life cycle impacts of the existing Irish housing stock. PhD thesis, Dublin Institute of Technology, Dublin. Ireland. Fuller, S and Petersen, S (1995) Life cycle Costing Manual for the Federal Energy Management Program, NIST Handbook, 1995 Edition. U.S. Department of Commerce, Washington DC. ISO (International Ofﬁce for Standardisation) (2008a) BS EN ISO 15686-5 Building and constructed assets – Service-life planning, Part 5 – Life Cycle Costing. ISO (International Ofﬁce for Standardisation) (2008b) BS EN ISO 15686-8 Building and constructed assets – Service-life planning, Part 8 – Reference service life and service Life estimation.

Daniel Coyle Paper.indd 13

08/11/2016 14:46

SDAR Journal 2016

List of abbreviations ach

Air changes per hour

AIRR

Adjusted Internal Rate of Return

BER

Building Energy Rating

BLCC5

Building Life Cycle Cost Program

CEN

European Committee for Standardisation

CO2

Carbon dioxide chemical formula

DEAP

Dwelling Energy Assessment Procedure

DHW

Domestic hot water

European Standard

EPBD

EnergyPerformance of Buildings Directive

European Union

EU-27

Total EU member countries

EWI

External wall insulation

ISO

International Organisation for Standardisation

LCC

Life Cycle Costs

LCCA

Life Cycle Cost Analysis

MVHR

Mechanical ventilation with heat recovery

NPV

Net Present Value

Net Savings

OM&R

Operation, Maintenance and Repair

Pascals (pressurisation units)

PHI

Passive House Institute

PHPP

Passive House Planning Package

Present Value

SEAI

Sustainable Energy Authority of Ireland

SCSI

Society of Chartered Surveyors Ireland

SHW

Solar hot water

Daniel Coyle Paper.indd 14

08/11/2016 14:46

Untitled-1 1

10/11/2016 08:53

BSN Arrow House Advert:Layout 1

26/10/2016

11:57

Page 1

Research Bank of Building Services Engineering The history of building services engineering in Ireland dates back to when our ancestors first began to research and devise engineering solutions to enhance their way of living. Archeological finds and studies have documented these down through the ages but, in respect of building services engineering, there is no single record dating back to then. However, there is a continuous record of all things relating to building services engineering in Ireland from April 1961 onwards, the year Building Services News was first published. Originally known as the Irish Heating & Plumbing Engineer, the journal has been in continuous publication since then and is now recognised as Irelandâ&#x20AC;&#x2122;s only dedicated building services journal. As such, Building Services News represents an invaluable treasure trove of building services engineering data and trends dating back over a continuous 55-year period. As a research data bank it is unique. That engineering research resource is now freely available for all to access as every single copy of Building Services News â&#x20AC;&#x201C; from April 1961 to the present day â&#x20AC;&#x201C; is currently being uploaded to the DIT Arrow site.

Log into http://arrow.dit.ie/bsn/

Find us on Facebook

Visit our website: buildingservicesnews.com

Hygrothermal risk evaluation IRU WKH UHWURÀ W RI D W\SLFDO solid-walled dwelling

Beñat Arregi BARCH (UPV/EHU)

BUILDING LIFE CONSULTANCY consult@buildinglifeconsultancy.com

Joseph Little MRIAI, MSC ARCHIT. AEES, BARCH (NUI) DUBLIN SCHOOL OF ARCHITECTURE, DIT joseph.little@dit.ie

Joe Little Paper.indd 1

08/11/2016 14:48

SDAR Journal 2016

Abstract

Note • WUFI Pro and WUFI 2D: A suite of tools developed by the

There is increasing evidence that current mainstream

Fraunhofer Institute for Building Physics since the early

guidance for assessing moisture risk of insulation retroﬁts

1990s. WUFI Pro is validated against EN 15026 (2007).

in Ireland and the UK is unsuitable for traditional solid-

WUFI 2D two-dimensional numerical simulation falls

walled buildings. This guidance is still based on simpliﬁed

outside the scope of the standard, but has been

hygrothermal risk assessment methods, despite the

repeatedly validated.

availability of more advanced numerical software for two decades and a relevant standard in place since 2007, EN 15026. Two-dimensional versions of these software applications can extend simulation beyond one-dimensional assemblies to more complex junctions. This exploratory study makes use of one of these advanced simulation tools, aided by physical measurement, to explore hygrothermal risks of solid wall retrofits at the junction with uninsulated and insulated ground floors. A brick-faced traditional dwelling in Dublin has been selected as a case study, and four scenarios have been simulated: its original condition and three retrofit approaches. Results indicate that (a) the moisture content at the base of the wall increases in all retrofit scenarios examined, and (b) the assemblies with high vapour permeability and no membranes result in the lowest hygrothermal risk. The findings should be supported by further research and could have great relevance to guidance, specification and grant policy for energy retrofits of solid wall properties in Ireland and the UK. Keywords: Energy-efficient retrofit; Vapour control layer; interstitial condensation; WUFI. Glossary: • Vapour control layer (VCL); • Damp-proof membrane (DPM).

Joe Little Paper.indd 2

08/11/2016 14:48

Hygrothermal risk evaluation for the retroﬁt of a typical solid-walled dwelling

1. Introduction In Ireland and the UK, the energy-efficient retrofit of the existing building stock plays an increasingly large role in meeting national targets for energy efficiency and carbon emissions. Current guidance provided to industry and homeowners for energy-efficient retrofit works in Ireland and the UK encourages the use of vapour control layers (VCL) and impervious materials such as damp-proof membranes (DPM), regardless of hygrothermal characteristics, orientation and location of existing structures(1,2) (see glossary). Indeed, grant aid can be contingent on following this guidance(3). This paper considers whether such guidance and grant aid is appropriate or misguided. The guidance has arisen due to the use of the Glaser method (and its antecedent, the dewpoint method) for many decades to consider hygrothermal risks by assessing the likelihood of condensation forming within building fabric assemblies. This simplified, steadystate calculation method, which assumes vapour diffusion is the only moisture transport mechanism(4), is repeated for each month using mean values. Given its limitations, it is most accurate when used to assess buildings in this climate where vapour from the room is the dominant source of moisture, e.g. low rise, airtight structures with ventilated, water-tight rainscreens(5): its results cannot be depended on in other cases. However, the method is still dominant in Ireland and the UK(6) and is frequently used to assess assemblies (such as solid brick masonry) that are outside the scope of its standard(7). This dominance appears to exist because of (a) the length of time

for which the method and its antecedent were the only assessment standard available to the UK and Irish construction industry, (b) its ease and speed of use, and (c) the inadequate referencing to EN 15026:2007(8) in BS 5250:2011(2) – the central document used in understanding and controlling condensation in British buildings. Ill-considered retrofits, or retrofit work specified after a risk assessment using an inappropriate method, can result in moisturerelated damages such as decay of bricks due to freeze-thaw, rot of timber joists, condensation in attics, or mould growth at cold surfaces, which is a potential health risk for occupants. With occupant health, building heritage and taxpayer’s money at stake it is essential low-risk retrofit strategies are undertaken based on sober evaluation under the appropriate standards. It is particularly inappropriate to use the Glaser method to assess brick or stone-faced traditional buildings, where liquid transport of wind-driven rain and ground moisture are typically of far greater significance than the transport of moisture via vapour diffusion. Numerical simulation tools, under the relevant standard(8), allow a much more accurate hygrothermal analysis of construction assemblies, by including all relevant moisture loads and transport mechanisms under realistic boundary conditions. The present study is an original contribution intended to extend recent research on the hygrothermal performance of traditional solid walls to their junction with ground floors, using two-dimensional transient numerical simulation software and limited physical measurement. Four different scenarios are compared: (a) the wall-ground junction as originally built; (b) in its current condition; (c) a mainstream retrofit using membranes as per current guidance; and (d) an alternative retrofit strategy based on vapour-permeable assemblies for both wall and ground floor.

2. Literature review In 2001 Pender(10) concluded that critical misconceptions of the moisture performance of solid walls were integrated into standard advisory practice. Despite the existence of solid ﬁndings from research, these had not become part of the common understanding in conservation circles or the construction industry. The International Energy Agency’s Annex 24 project(9) reported that (a) airtightness is the most important performance requirement, (b) vapour diffusion from the room only poses a threat in absence of airtightness or for severe indoor climates, and (c) a vapour retarder may prevent drying of built-in moisture.

Figure 1 – Case study house in Ranelagh, Dublin.

The convenience of vapour control layers for internal insulation of solid walls in Continental Europe has been challenged by many independent studies in Germany(11), Denmark(12), Belgium(13) and Switzerland(14). Hygrothermal risk assessments of traditional solid walls in the climates of Ireland(15) and Scotland(6), carried out by these authors using transient numerical simulation, also concluded that (a) preserving drying capacity is more critical than preventing vapour

Joe Little Paper.indd 3

12/11/2016 09:53

SDAR Journal 2016

preserving drying capacity is more critical than preventing vapour ingress from the room, and (b) the addition of impervious layers (as recommended by mainstream guidance) can result in moisture accumulation. Similar findings have been reported from physical measurement in a recent field study in Dublin(16). In contrast with the growing number of empirical and desktop studies on insulation retrofits of walls, the volume of research on ground floors remains very limited, possibly due to their greater complexity to simulate and measure.

3. Case study A two-storey terraced house in Ranelagh, Dublin (Figure 1) has been selected as a case study. It is a modest-sized, mid-terrace Edwardian red brick dwelling, characteristic of the beginning of the 20th Century. The front wall is laid in solid brick masonry featuring a Flemish bond. Its original lime pointing was replaced with sand-cement jointing in the 1970s and internal lime plastering was replaced by tanking in the mid-1990s. According to the owner, the edges of bricks have been spalling for many years at interface with mortar joints and, more recently, whole bricks have lost, or are losing, their facing (see damage in Figure 2). Spalling is usually caused by the mechanical action of water (freezing and thawing) stressing the pore structure of the masonry unit: it is a clear indicator of hygrothermal stress. Four scenarios have been modelled for the wall-ground junction of the case study house. These are detailed below:

(a) Original condition. The house as originally built circa 1905: •

Lime mortar joints with lime pointing;

•

Lime plaster as internal ﬁnish;

•

Suspended timber ﬂoor with ventilated underﬂoor space;

•

No damp-proof membranes.

(b) Current condition. After retroﬁt carried out in the mid-1990s, following what were then understood to be best practice measures: •

Sand-cement jointing over lime mortar;

•

Sand-cement render with waterprooﬁng admixture applied internally, with skim plaster ﬁnish;

•

Concrete ﬂoor over damp-proof membrane and expanded polystyrene insulation.

(c) Mainstream retroﬁt. A likely low-cost insulation retroﬁt to the wall, following current mainstream guidance and attracting grant aid: •

Internal wall insulation to U = 0.27 W/m²K: composite boards with polyisocyanurate (PIR) insulation, vapour-closed foil facing and plasterboard ﬁnish;

•

No works in ground ﬂoor over current condition.

(d) Proposed retroﬁt. An alternative approach encouraging free transport and dissipation of moisture: •

Removal of cement jointing and repointing with lime;

•

Removal of internal waterprooﬁng render and re-application of lime plaster;

•

Internal wall insulation to U = 0.60 W/m²K: calcium silicate bonded to wall with lime-based adhesive and lime-based plaster ﬁnish;

•

New lime screed flooring slab over insulating layer of recycled foamed glass aggregate (in lieu of original suspended floor or later concrete floor with DPM); materials and specification broadly based on (17);

•

No vapour control layers or damp-proof membranes.

4. Methodology The hygrothermal performance of the four scenarios described above has been numerically simulated using WUFI software(18) developed by the Fraunhofer IBP (see glossary). Given that a junction of wall with ﬂoor is assessed, involving two-dimensional heat and moisture ﬂows, the variants in this case study have been modelled using WUFI 2D v3.4. This software has been experimentally validated numerous times, including through the simulation and measurement of the two-dimensional effects of rising damp(19).

Figure 2 – Damage to brick in case study house: (above) whole brick spalling, (below) loss of brick edge and surface at base of external wall.

The external climate data (a reference year with hourly inputs including driving rain) has been generated using Meteonorm v6.1(20), based on interpolated weather data for Dublin Airport in accordance with the procedure in the standard(8). The rainwater exposure model within WUFI(21) for buildings up to 10m high has been applied. The internal climate is based on a normal moisture load(8) as a function of external climate data, resulting in an indoor relative humidity range of 40–60%. For the ground boundary condition, a constant relative

Joe Little Paper.indd 4

08/11/2016 14:48

Hygrothermal risk evaluation for the retroﬁt of a typical solid-walled dwelling

humidity of 99% has been assumed, with temperatures defined by a sine curve fluctuating between 5°C and 15°C (an extrapolation of ground measurements by the Fraunhofer IBP in Holzkirchen, Bavaria, to the Dublin climate). The two-dimensional models built for the simulations are depicted in Figure 3. Due to the significant computational time involved, the duration of the simulations has been limited to three years (26,280 hourly calculations), starting on 1st October as per usual convention. (Note: a three-year duration was judged acceptable after a longer one-dimensional assessment using WUFI Pro.)

Figure 4 – Water absorption measured for wall of case study house, compared to other walls measured in Dublin.

The readings obtained from Karsten tubes have been converted into a water absorption coefﬁcient(6), a measure of one-dimensional water uptake over square root of time, to allow use within the WUFI software. The A-value obtained is 0.18 kg/m²3s. The moisture sorption characteristics of a given material are described by its moisture storage function, which indicates the equilibrium water content of the material as a function of relative humidity (see glossary). The moisture storage function is measured in laboratories using sorption isotherms and pressure plate measurements of a material sample(22). Approximate values can also be simply measured by suspending materials above salt solutions (of known relative humidity) but require destructive testing which was not possible in this case.

(1) brick, (2) ventilated air layer, (3) softwood floor, (4) lime plaster, (5) DPM, (6) EPS insulation, (7) concrete slab, (8) waterproof cement plaster, (9) PIR composite plasterboard, (10) compacted aggregate of foamed glass, (11) cork edge insulation, (12) lime screed, (13) calcium silicate insulation. • Reference points for relative humidity. Figure 3 – Two-dimensional models for wall-ground junction, clockwise from top left: (a) original condition; (b) current condition; (c) mainstream retrofit; (d) proposed retrofit.

5. Material properties When determining the hygrothermal performance of a solid masonry wall, the most critical properties of the substrate are its moisture absorption and storage characteristics(6). The water absorption of the wall has been measured by an insitu, non-invasive, test using ten Karsten tubes (Figure 2 bottom), following RILEM Test Method II.4, in which the imbibed amount of water is measured at regular time intervals(6). The assessed wall sits in a mid-low range of absorption, when compared to other brick walls measured by the authors in Dublin (Figure 4): that is to say it is far more absorbent than rendered walls measured but less absorbent than the mean absorption rate for brick walls measured to date.

Figure 5 – (Top) Moisture storage function of bricks in the MASEA database that match the measured water absorption of the case study wall, (bottom) a scale where colour is used to graphically indicate level of water content present.

Figure 5 plots the moisture storage function of six bricks in the MASEA database(23) that match the water absorption coefﬁcient measured for the case study house. For the purpose of this study one of the bricks (Solid Brick ZL) has been selected as its moisture storage characteristic is representative (see red line in Figure 5). As can be seen in Figure 5, the increase in moisture content is relatively

Joe Little Paper.indd 5

08/11/2016 14:48

SDAR Journal 2016

steady and below 95% relative humidity (driven by diffusion and absorption of water as vapour) but increases dramatically above that threshold as capillary ﬂow of liquid water becomes dominant. The scale below the graph conveys this signiﬁcance through colour. Table 1 lists all the materials used in the two-dimensional models with their most relevant hygrothermal characteristics. Moisture storage functions are indicated by the reference water content (w80) and the

A, water absorption coefﬁcient (kg/m²3s)

wf, free water saturation (kg/m³)

Wall substrate

w80, reference water content (kg/m³)

μ, vapour diffusion resistance factor

Table 1 – Material data for simulated scenarios

Solid Brick ZL(1)

5.5

216

0.183

Lime mortar(3)

6.5

248

0.153

free water saturation (wf), corresponding to equilibrium moisture contents at 80% and 100% relative humidity, respectively. The floor assembly of proposed retrofit (d) is the “Sublime Insulated Limecrete Floor” supplied by Ty^-Mawr Lime Ltd in Wales. This system has approval from LABC and LABSS bodies in the UK. The insulating hardcore that forms the most novel part is recycled foamed glass manufactured in Germany by GLAPOR (see Figure 6). This highlyvapour-permeable layer which acts as floor base, capillary break and insulant (Ï= 0.078 W/mK), is compacted on site to ~75% of its initial height. The water content values listed in Table 1 for this material are estimates based on comparisons with other granular materials. Except in areas of high radon, the manufacturers recommend that the only membranes below and above the hardcore are geotextiles, so as to ensure a fully-vapour-permeable assembly.

(a) Original condition Lime plaster(2)

250

0.050

200

575

–

0.07

–

25.7

210

0.057

Cement plaster with waterproof admixture

750

280

0.008

Skim plaster(2)

8.3

6.3

400

0.287

100m

–

EPS insulation

–

Concrete

180

150

0.003

Softwood(1) Underﬂoor air layer

(3)

(b) Current condition Cement pointing(5) (5)

DPM * (5)

(1)

0.73

–

Foil facing(4)*

20m

–

Plasterboard

8.3

6.3

400

0.287

Lime-based adhesive(4)

22.9

280

0.004

Calcium silicate board

5.4

7.1

815

0.930

Lime-based plaster(4)

250

0.047

Insulating hardcore

–

Lime screed

152

0.016

Cork edge insulation(2)

–

PIR insulation(4) (2)

(d) Proposed retroﬁt (4)

(5)

Source of data: (1) MASEA database in WUFI; (2) Fraunhofer IBP database in WUFI; (3) other databases in WUFI; (4) manufacturer data; (5) adapted by assessor * Vapour resistance given as sd value.

For reasons of space values less pertinent to a discussion about moisture are not listed here.)

Figure 6 – Construction worker compacting recycled foamed glass aggregate during installation of “Sublime Insulated Limecrete Floor” (image supplied by Ty^-Mawr Lime Ltd).

6. Assessment criteria Moisture, warmth, oxygen and nutrients are all necessary ingredients for mould growth(24). Mould is generally prevented from forming on the internal surfaces of buildings by ensuring that relative humidity on those surfaces is maintained below 80%(2). Within the layers of a building component it has also been long accepted that relative humidity, should not exceed 80% for sustained periods, where temperatures are sufﬁciently high to support mould growth and a potential exists for mould to affect occupants. Where unintended air paths allow the interchange of air between the room and a void behind insulation (e.g. through gaps under the skirting or at a pattress box) it seems logical to apply the same threshold value;

Joe Little Paper.indd 6

08/11/2016 14:48

Hygrothermal risk evaluation for the retroﬁt of a typical solid-walled dwelling

but this threshold may not be as relevant where there is no void or the materials present inhibit mould growth. Internal wall insulation systems that are fully bonded to the wall should provide additional safety, especially if the adhesive, and in some cases the insulant, are alkaline (as they can act as biocides, i.e. mould suppressants). This suggests that a higher risk assessment threshold than 80% relative humidity could make sense in those cases. The WTA (a European transnational scientiﬁc-technical association involved in the development of standards)(25) has in fact reported that if the potential for mould growth is removed, the acceptable interstitial relative humidity threshold for certain assemblies may be shifted upwards to 95%, as the risk of material decay and freezethaw damage become the key concerns(26). Given the increasing importance of energy-focused retroﬁt work and the need to ensure low-risk interventions, it is advisable that the WTA

research and the appropriateness of using a higher relative humidity threshold than 80% for certain conditions be studied for applicability to retroﬁt work in the UK and Ireland.

7. Evaluation of results Figure 7 makes use of a coloured scale to portray the distribution of relative humidity over the wall-ground junction, for the four assemblies simulated (each in a different column), at three particular moments during the simulation. These moments are a rain event (top diagrams), a drying-out period (middle diagrams), and a relatively dry period (bottom diagrams), allowing comparison of the relative performance of the four assessed scenarios. The correlation between relative humidity and water content is critical for assessing moisture-related risks such as mould growth

Figure 7 – Distribution of relative humidity in four numerical simulations with coloured scale. Columns, left to right: (a) original condition, (b) existing condition, (c) mainstream retroﬁt, (d) proposed retroﬁt. Within each column: during a rain event (top), during the drying-out process (middle), during a relatively dry period (bottom).

Joe Little Paper.indd 7

08/11/2016 14:48

SDAR Journal 2016

or brick spalling. This is especially so above 95% relative humidity, where the quantities of water involved rise dramatically (Figure 5). The significance of this change has been conveyed in the scale for relative humidity in Figure 7, where every percentage point above 95% is identified by a unique shade of blue. Driving rain tends to cause a sudden increase in the moisture content of outer part of the wall substrate (top diagrams in Figure 7). After the rain event, the outer face of the wall dries out first (middle diagrams in Figure 7), while part of the absorbed moisture migrates towards the core of the wall. Note that, in every case, the lowest brick courses show higher moisture content (blue and dark blue tones in Figure 7) due to capillary absorption of moisture from the ground(27). In the original condition of the wall (column (a) in Figure 7), the absorbed moisture is freely transported to the inner and outer surfaces and evaporates to the air volumes on each side. In a well-built and well-maintained traditional building some level of equilibrium is reached, wherein such cycles occur annually without negative impact on building fabric or occupants. Similarly, any moisture at the base of the wall can dry outwards or inwards to the ventilated underfloor space below the suspended floor, thereby keeping the internal surfaces (lime plaster and timber flooring) at safe relative humidity levels (i.e. below 80%). The current condition (column (b) in Figure 7) shows a noticeable increase in humidity within the lowest brick courses. This is caused by (1) a drop in temperature in this area of the wall after the addition of floor insulation and (2) the removal of the ventilated underfloor that allowed local evaporation from the rising wall. Note how the DPM between the wall substrate and the floor insulation prevents the passage of vapour and creates a build-up of condensation (black colour in Figure 7) between floor insulation and membrane. Above the floor, there is also an increase in the overall moisture content of the wall, due to the waterproofing plaster that inhibits the moisture in the wall from drying out towards the room. These increases in water content might be related to the observed brick spalling (Figure 2). For the mainstream retrofit (column (c) in Figure 7), the masonry substrate remains consistently moist at the junction with the insulation. The accumulated moisture cannot dry towards the room due to the presence of vapour resistant materials (the waterproofing plaster and the foil facing within the PIR insulation). As composite insulated plasterboard systems of this kind feature a cavity behind the insulation (due to the use of dabbing or studs), which is likely to be linked to the room through unintended air paths, mould growth at the internal face of the wall substrate behind the internal insulation should be considered a specific risk. The floor assembly is unchanged from condition (b), but due to the overall increase of moisture in the wall, the thin black zone indicating 100% relative humidity and greatest water content grows higher, reaching internal floor level. This condensate could remain hidden to the view, leading to slow degradation of adjacent materials, or could manifest internally as a source of moisture below the skirting. The proposed retrofit (column (d) in Figure 7) features a fully-bonded capillary-active internal insulation system (calcium silicate boards) and vapour-permeable flooring (lime screed) over an insulating

Figure 8 – Evolution of relative humidity in numerical simulations, at inner face of brick above internal ﬂoor level (black dots in Figure 3), for the four scenarios assessed.

capillary break of foamed glass aggregate(17). This approach results in significantly lower relative humidity levels than the mainstream retrofit scenario (c), and all areas in the vicinity of internal surfaces remain uniformly dry (red colour in Figure 7). It is therefore considered to entail lower risk of mould growth and material decay. Figure 8 compares relative humidity levels for the four scenarios over the length of the simulation. The internal edge of the brick above internal floor level (black dots in Figure 3) has been chosen as an important location to study as it is simultaneously affected by conditions at wall and ground floor, and is a sensitive location due to its proximity to the room surface. In the context of the moisture storage function of the selected brick (red line in Figure 5), it is apparent from Figure 8 that: •

The original condition (a) has the lowest relative humidity and thus lowest moisture content. It also displays a clear seasonal nature, averaging less than 75% RH.

•

The current condition (b) results in greatly increased humidity at the assessed location averaging 97% RH with little drying effect: this shows the base of the wall is already hygrothermally stressed, as can be seen in Figure 2.

•

The mainstream retrofit (c) results in an increase that on average is only 0.8% higher than (b) but exhibits no drying effect and peaks extending to 99.5%. In the context of the marked increase in moisture content after 95% RH and even more so 97% RH (Figure 5), the associated stressing is significant. The moisturerelated risk of this retrofit should be considered unacceptable.

•

The proposed retrofit (d) results in lower humidity levels (averaging ~93.8% RH) than conditions (b) and (c) with the absence of the damp-proof membrane having a beneficial effect on local drying. While (d) has a far higher relative humidity at the measurement point than (a), it does exhibit a summer drying effect that ensures the overall performance is out of the vulnerable zone above 95% where pores are increasingly filled and capillary action dominates moisture transfer. This will help protect the masonry, reducing the potential for spalling of the wall’s outer surface. As the location is behind alkaline lime plaster and a fully-bonded internal wall insulation assembly, the likelihood of mould forming is low and the risk to occupants negligible.

Joe Little Paper.indd 8

08/11/2016 14:48

Hygrothermal risk evaluation for the retroﬁt of a typical solid-walled dwelling

In Table 2, results are summarised in the form of a matrix, considering both thermal insulation (x axis) and associated hygrothermal risk within the wall substrate (y axis). The original condition (a) has the lowest risk of damage to the wall substrate, at the cost of a poor thermal performance. If an insulation retrofit is to be carried out, the proposed retrofit (d) is the safest option according to this risk assessment using transient numerical simulations. It should be borne in mind that any internal wall insulation retrofit will compromise the drying ability of the wall substrate as it isolates it from the internal heating system. This effect is amplified as internal insulation levels increase(6). Insulants that are vapour closed such as used in (c) will also limit drying to the room. The current condition (b) stresses the masonry substrate because drying to the room is inhibited by tanking, even though the wall surface is heated by the dwelling’s heating system. Retrofit condition (c) adds to the risk by also thermally isolating the substrate, while (d) thermally isolates it but allows vapour and capillary movement in both directions. no insulation high risk

ﬂoor insulation

wall & ﬂoor insulation

current condition

mainstream retroﬁt proposed retroﬁt

moderate risk low risk

•

Ideally the dwelling would also feature mechanical extract ventilation that constantly removes indoor air contaminants, as an indoor air quality measure;

•

In high radon areas the suppliers of the ﬂoor system acknowledge that a radon barrier should be used. (The role of mechanical extract ventilation systems in managing radon in low and medium radon areas falls outside the scope of this paper)

This study shows clearly the kind of risk assessment possible using transient numerical simulation like WUFI Pro and 2D. This is not possible for a wide range of reasons with the Glaser method as discussed. The study raises serious questions about current construction practice, mainstream guidance and grant aid policy in Ireland and the UK, especially where a mainstream approach appears to increase hygrothermal risks to historic dwellings. There is a need for parametric modelling to expand this assessment to a range of wall and ground assemblies, insulants and locations: ideally, this would be supported by selected physical testing. Acknowledgements This research work has been carried out within the Built to Last project of Dublin City Council (part-funded by Heritage Council and Department of Arts Heritage and Gaeltacht). The Project explores strategies for the energy-efﬁcient retroﬁt of pre-1945 historic dwellings in Dublin.

original condition

Table 2 – Result matrix indicating insulation (x axis) and hygrothermal risk within wall substrate (y axis).

8. Discussion/conclusion In this study, the hygrothermal performance of four different scenarios (original condition and three retrofit approaches) has been assessed for the ground junction of a traditional brick-faced house, using two-dimensional numerical simulation. While limited to one case study house in the Dublin climate, the findings appear to indicate that vapour permeable assemblies should be favoured for insulating ground floors and external walls of brick-faced solid wall buildings. The use of damp-proof and vapour-resistant membranes appears to result in higher moisture content within the wall and ground slab. These findings are consistent with recent research on the hygrothermal performance of walls(6,16) but contradict current guidance(1,2), which is based on simplified assessment methods that are unsuitable for assessing traditional buildings(5,7). While a choice of vapour-permeable insulants and strategies are now available for internal wall insulation, the range of products for floors remains much more limited. The suitability of this ground floor assembly without a damp-proof membrane is contingent on the following: •

The ﬂoor insulation should be designed to act as a capillary break preventing rise of ground moisture;

Joe Little Paper.indd 9

08/11/2016 14:48

SDAR Journal 2016

References (1)

Technical Guidance Document L (2011): Conservation of Fuel and Energy – Dwellings. Government of Ireland.

(2)

BS 5250 (2011): Code of practice for control of condensation in buildings. BSI.

(3)

Better Energy Scheme Contractors Code of Practice and Standards and Speciﬁcations Guideline (6th ed.) (2013). Dublin: SEAI.

(4)

Rose, W. (2003): The rise of the diffusion paradigm in the US. In: Carmeliet et al., Research in Building Physics, p. 327–334.

(5)

Hens, H. (2011): Applied Building Physics. Berlin: Ernst & Sohn.

(6)

Little, J., Ferraro, C. and Arregi, B. (2015): Technical Paper 15 – Assessing risks in insulation retroﬁts using hygrothermal software tools. Edinburgh: Historic Environment Scotland.

(7)

I.S. EN ISO 13788 (2012): Hygrothermal Performance of Building Components and Building Elements – Internal Surface Temperature to Avoid Critical Surface Humidity and Interstitial Condensation – Calculation Methods. NSAI.

(8)

I.S. EN 15026 (2007): Hygrothermal Performance of Building Components and Building Elements – Assessment of Moisture Transfer by Numerical Simulation. NSAI.

(9)

Hens, H. (1996):. Heat, air and moisture transfer in insulated envelope parts: Volume 1: Modelling. Final report. (IEA Annex, 24) Leuven, Belgium: Katolieke Universiteiet Leuven, in association with International Energy Agency.

(23) MASEA: Material database for energy-efficient retrofit. (Online database available at http://www.masea-ensan.com) (24) Sedlbauer, K. (2001): Prediction of mould fungus formation on the surface of and inside building components, PhD Thesis, Fraunhofer IBP. (25) WTA: International Association for Science and Technology of Building Maintenance and Monuments Preservation. (http://www.wta-international. org) (26) WTA Merkblatt 6-4 (2009): Innendämmung nach WTA I: Planungsleitfaden. Wissen-schaftlich-Technische Arbeitsgemeinschaft für Bauwerkserhaltung und Denkmalpflege e.V. (27) Hall, C. & Hoff, W.D. (2007): Rising damp: Capillary rise dynamics in walls. In: Proceedings of the Royal Society A ¬– Mathematical, Physical and Engineering Sciences, Vol. 463, p. 1871–1884.

(10) Pender, R. (2001): The behaviour of moisture in the porous support materials of wall paintings and investigation of some environmental parameters. Ph.D thesis, Courtauld Institute of Art (University of London). (11) Worch, A. (2010): Innendämmung: Bauphysikalische Aspekte, Probleme und Grenzen: Lösungswege für die Praxis. In: 8. Allgäuer Baufachkongress, Oberstdorf, Germany. (12) Nielsen, A. et al. (2012): Use of sensitivity analysis to evaluate hygrothermal conditions in solid brick walls with interior insulation. In: Proceedings of the 5th International Building Physics Conference (IBPC). (13) Evrard, A. et al. (2014): Influence of liquid absorption coefficient on hygrothermal behaviour of an existing brick wall with lime-hemp plaster. In: Building and Environment 79, p. 90–100. (14) Guizzardi, M. et al. (2015): Risk analysis of biodeterioration of wooden beams embedded in internally insulated masonry walls. In: Construction and Building Materials 99, p. 159–168. (15) Little, J. and Arregi, B. (2013): Managing moisture – the key to healthy internal wall insulation retrofits of solid walls. In: Proceedings of 17th International Passive House Conference, Frankfurt. (16) Walker, R. and Pavía, S. (2016): Performance of a range of thermal insulations in a historic building. In: Engineers Journal. Published online on 22 March 2016. (17) Sublime® Limecrete floor, available from T -Mawr, Wales. (https://www. lime.org.uk) (18) WUFI®: Heat and Moisture Transient Simulation. (http://wufi.de/en) (19) Holm, A. and Künzel, H.M. (2000): Two-dimensional transient heat and moisture simulations of rising damp with WUFI 2D. Fraunhofer IBP. 12th International Brick and Block Masonry Conference, Madrid, Spain. (20) Meteonorm. (http://www.meteonorm.com) (21) Straube, J.F. (1998): Moisture Control and Enclosure Wall Systems, PhD Thesis, Civil Engineering Department, University of Waterloo. (22) Krus, M. (1996): Moisture Transport and Storage Coefficients of Porous Mineral Building Materials. Theoretical Principles and New Test Methods. Fraunhofer IRB Verlag.

Joe Little Paper.indd 10

08/11/2016 14:48

SDAR Awards Abstracts 2017:Layout 1

26/10/2016

11:53

Page 1

Call for Abstracts Short abstracts (between 100/200 words max) for entry into the SDAR Awards 2017 must be submitted by mid-January 2017, by email directly to Michael McDonald and/or Kevin Kelly of DIT at michael.mcdonald@dit.ie and kevin.kelly@dit.ie The SDAR Awards is a joint initiative between CIBSE Ireland and DIT, supported by Building Services News, and sponsored by John Sisk & Son. The awards are unique in that they are intended to disseminate knowledge, encourage research in sustainable design of the built environment and raise the quality of innovation and evaluation of 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. This competition is open to architects, engineers and all professionals involved in construction projects. Now more than ever as positive signs ripple through the built environment, this unique synergy between industry and academia allows greater potential for integration of modern low-carbon technologies and low-energy design methodologies. The SDAR Awards competition is intended to create a platform for the growth of applied research in the expanding green economy. Postoccupancy evaluations and similar critical appraisal of low-energy projects facilitate the transition from ideologically-driven innovations, sometimes offering poor value, to evidence-based applied research that proves value or identifies

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

weaknesses that the industry can learn from. These successes and failures help inform the professional community across all the building industry disciplines. From the abstracts submitted by the mid-January 2017 deadline, a shortlist will be selected by peer review, and those selected will be invited to prepare and submit final papers by March 2017. The final will take place in April 2017 in DIT, Bolton Street. First prize is a cheque for â&#x201A;Ź1000. Candidates that present at the awards also have a chance of publishing their papers in the SDAR Journal â&#x20AC;&#x201C; arrow.dit.ie/sdar/ For further information contact: michael.mcdonald@dit.ie or kevin.kelly@dit.ie

Sponsored by :

Breathing Life into Buildings a career that is exciting, makes a real

Building Services Engineers …

difference to people’s lives and safeguards

•

Have a passport to international work

•

Make a social and economic difference

•

Bring water to a village in Africa

•

Design a data centre for Google

•

Beam energy from solar space stations

•

Computer-simulate environmental events

•

Enhance people’s lives

Don’t settle for the ordinary – choose

the future of the planet … Become a Building Services Engineer. Architects design the body, civils build the skeleton but it is building services engineers who breathe life into buildings. They are the ones who provide the heart, lungs and veins of every building we work, play and live in!

www.cibseireland.org CIBSE 'Why Choose' Advert 2016.indd 1

14/11/2016 09:21

Enhancing Thermal Mass Performance of Concrete

Assessment of two methods of enhancing thermal mass performance of concrete through the incorporation of phase-change materials

Dervilla Niall

DUBLIN INSTITUTE OF TECHNOLOGY, IRELAND AND TRINITY COLLEGE, UNIVERSITY OF DUBLIN, IRELAND dervilla.niall@dit.ie

Roger P. West

TRINITY COLLEGE, UNIVERSITY OF DUBLIN, IRELAND

Sarah McCormack

TRINITY COLLEGE, UNIVERSITY OF DUBLIN, IRELAND

Niall, West McCormack Paper.indd 1

14/11/2016 10:01

SDAR Journal 2016

Abstract According to the IEA Technology Roadmap on Energy Efﬁcient Building Envelopes, buildings are responsible for more than one third of global energy consumption, with space heating and cooling consuming 33% of this energy, and increasing to 50% in cold climates. Using the mass of a building to store heat and/ or cold can reduce the demand on the auxillary heating and/or cooling systems and hence reduce the overall energy demand of the building. In this study the thermal storage capacity of concrete was actively enhanced by integrating phase-change

1. Introduction 1.1 Background theory The use of thermal energy storage systems (TES) is recognised as one of the most effective approaches to reducing the energy consumption of buildings. One of the main issues with renewable energy sources such as solar is that the supply is intermittent. A TES system can be used to absorb and store both solar energy and excess heat due to use and occupancy during the day, which can then be released to the internal environment when the room temperatures fall at night. In this way a TES system provides a potential for improved indoor thermal comfort for occupants by moderating internal temperature ﬂuctuations and reducing the overall energy consumption of the building due to load reduction and shifting electricity consumption to off-peak periods (Figure 1).

materials (PCMs) which provide a high latent heat storage capacity. Two methods of incorporating PCMs into concrete were used to form PCM/concrete composite panels. The ﬁrst type of panel was formed by adding microencapsulated parafﬁn to fresh concrete during the mixing process. The second panel was formed by vacuum impregnating butyl stearate into lightweight aggregate which was then included in the concrete mix. The aim of the study was to compare the thermal behaviour of both PCM/concrete composite panels to a control concrete panel and to evaluate which method of PCM incorporation is the most effective at improving thermal mass characteristics in the context of a thermal energy

Figure 1. Stabilising effect of thermal mass on internal temperatures [1]

lightweight aggregate/PCM composite is more effective at

A TES system can include sensible heat storage, latent heat storage or a combination of both. Sensible heat storage systems store energy by increasing the temperature of the material. The capacity of a material to store energy depends on the amount of energy required to change the temperature of a unit amount of the material, ie, the speciﬁc heat capacity of the material. The storage capacity of a sensible heat system is given by [2]:

providing additional thermal storage, particularly within the

Q = ∫Tif mCpdT (1)

ﬁrst 100mm of depth of an element of structure.

where

Keywords:

Q = quantity of thermal energy stored, (Joules). Tf & Ti = ﬁnal temperature and initial temperature respectively (oC). m = mass of material. Cp=speciﬁc heat capacity of material (J/kgK).

storage system for space heating/cooling in a building. The panels containing PCM displayed signiﬁcantly greater thermal storage capacity, despite having reduced thermal conductivity and density. The study concluded that the panel containing

Phase Change Materials (PCMs), PCM/concrete composite, Thermal diffusivity, Thermal storage.

Using the mass of a building as a TES system is an established approach in the design of energy efficient buildings and is commonly referred to as thermal mass. Thermal mass = mass x specific heat capacity (2) One of the main parameters that influence thermal mass behaviour is thermal diffusivity, α, which is the ratio of the conductivity of a material to its volumetric heat storage capacity given by the equation:

Niall, West McCormack Paper.indd 2

14/11/2016 10:01

Assessment of two methods of enhancing thermal mass performance of concrete through the incorporation of phase-change materials

α=

k ρCp

(m2/s) (3)

where Ú is the density (kg/m3), k is the thermal conductivity(W/ mK) and Cp is the speciﬁc heat capacity (J/kgK). Thermal diffusivity indicates the rate at which temperature changes occur in a material. The higher the value of thermal diffusivity the quicker the material will reach temperature equilibrium with its environment. Another key property that inﬂuences thermal mass behaviour is the thermal inertia of a material denoted ‘I’ which is a measure of the responsiveness of a material to variations in temperature. Thermal inertia is given by the following equation [3]: I = √ρCpk (J/(m2K√s) (4) A high thermal inertia describes materials that characterise high thermal mass and high thermal conductivity. Such materials will display small changes in temperature throughout the diurnal cycle. Referring to equation (3) for thermal diffusivity, α, equation (4) can also be written as follows: I=

k √б

(J/(m2√s) (5)

It can be noted from equation (5) that the higher the thermal diffusivity of a material the lower the thermal inertia. Hence for a building material to provide good thermal mass it requires an appropriate balance between thermal diffusivity and thermal inertia. Concrete is a building material that combines a high specific heat capacity and high density with a thermal conductivity that is appropriate for the diurnal heating and cooling cycle of buildings and hence has good thermal mass characteristics. In this study the thermal storage capacity of concrete was enhanced by adding phasechange materials (PCMs) which provide a high latent heat storage capacity. A PCM is a material that absorbs high amounts of thermal energy while changing phase, ie, solid to liquid and liquid to gas. The change in temperature of PCMs during phase-change is insignificant. When incorporating PCMs into construction materials it is only the liquid-solid phase-change that is utilised. As temperature increases the PCM absorbs the heat and changes phase from solid to liquid. When the temperature decreases the PCM releases the heat as it changes from liquid to solid. In this way PCMs can be used to control air temperatures within a building [4]. Kosny et al [5] carried out an overview of the potential applications of phase-change materials in building envelopes which depend largely on local climate conditions and melt temperature range of the PCM. The overall thermal capacity of a PCM/concrete composite is a combination of the specific heat capacity of the concrete, the specific heat capacity of the PCM and the latent heat capacity of the PCM. It will vary depending on the state of the phase transition of the PCM. The total energy stored by a PCM composite, omitting a liquid to gas phase-change, can be written as: T

T T ∫T fmcCpcdT + ∫T mi mpcmCpsdT+ mpcmL + ∫Ti fmpcmCpldT (6) i

where Qpcm-comp = overall thermal capacity of PCM/concrete composite Tmf & Tmi = final melt temperature and initial melt temperature respectively (oC). mc & mpcm = mass of concrete and mass of PCM respectively (kg). Cpc = specific heat capacity of concrete (J/kgK) Cps & Cpl = specific heat capacity of solid PCM and liquid PCM respectively (J/kgK) L = latent heat capacity of PCM (J/kg) 1.2 Selection of PCM There are many different types of PCMs which can be broadly classified into three categories based on their chemical composition – organic, inorganic and eutectics. Pasupathy et al (2008) [6] summarised the advantages and disadvantages of each category. Organic PCMs mainly comprise paraffin and fatty acids and are compatible with most construction materials. Paraffin wax is a hydrocarbon with a chemical structure CnH2n+2. Paraffins typically have melting temperatures ranging between 20oC and 70oC. The higher the number of carbon atoms in the chain the higher the melting temperature [2]. A number of researchers, ( [7], [8], [9], [10], [11] and [12] ) have carried out thermal energy storage studies that combined paraffin with concrete. The paraffin is micro-encapsulated in thin polymer shells which are then added to fresh concrete towards the end of the mixing process. The microcapsules provide a large surface area for heat transfer and also resist volume change during the phase transition. Cabeza et al.[11] constructed two full size cubicles with windows incorporated in each wall, one with ordinary concrete and the other with a concrete which contained 5% by weight micro-encapsulated paraffin. The study successfully demonstrated an increase in thermal storage capacity and increase in thermal inertia for the PCM/concrete composite. Generally from the review of studies that considered PCM/concrete composites, paraffin appears to be the most common choice of PCM as it is inactive in an alkaline medium, chemically stable and relatively inexpensive. However, paraffin has a relatively low conductivity and the capsules also adversely affect the mechanical properties of the concrete [13]. Fatty acids have melting temperatures similar to that of paraffin. They also have good melting and freezing properties. However, they are generally more expensive than paraffin [13]. Butyl stearate is a less expensive fatty acid which has also been successfully combined with concrete in previous research ( [14] and [15] ). Zhang et al [15]. compared different methods of incorporating PCMs into porous aggregates and concluded that the vacuum impregnation method was the most effective. In the vacuum impregnation method the air was evacuated from the porous aggregate using a vacuum pump. The aggregates are then soaked in a liquid PCM under vacuum. The porous aggregate was then added to a concrete mix. The thermal energy storage capacity of the concrete samples were assessed using differential scanning calorimeter tests and it was found that the energy absorbed by the PCM/concrete composite samples increased almost linearly with the volume fraction of PCM in the concrete.

Niall, West McCormack Paper.indd 3

14/11/2016 10:01

SDAR Journal 2016

Inorganic phase-change materials are salt hydrates. Salt hydrates are alloys of inorganic salts and water. They have high latent heats per unit mass and volume. They also have high conductivity, almost double that of paraffin [2]. However, salt hydrates have two particular problems – incongruent melting and supercooling – which reduces the efficiency of the heat transfer. A eutectic mixture is a combination of chemical compounds or elements that , at a particular combination, will solidify at a lower temperature than any other combination. Eutectics nearly always melt and freeze without segregation since they freeze to an intimate mixture of crystals. Also on melting, both components liquefy simultaneously. However, there is only limited data available on the thermo-physical properties of eutectics as the use of these materials is still new to thermal storage applications. Primarily, the selection of a PCM should ensure that the melt temperature range of the PCM is suitable for the intended application. For a space heating application in a building, phase-change materials with a melting temperature within the range of human comfort temperature (18-22oC) are suitable [6]. For a space cooling application the appropriate melt temperature range is higher at 19 - 24oC [16]. The phase-change material selected must also be chemically compatible with the material in which it is to be incorporated. For this study organic PCMs were deemed the most suitable for incorporating into concrete. Hence, taking into account the suitable melt temperature range for a space heating/cooling application, paraffin and butyl stearate were selected. 1.3 Objective of study Previous research has been carried out on many types of PCM/ concrete composites. However, the effectiveness and performance of the different types of composites have not been compared in previous studies. The depth of heat penetration into a PCM/concrete panel depends on the thermal conductivity of the panel, the latent heat of the PCM and the temperature of the internal environment in which the panel is located. The PCM in the panel will only be effective, that is change phase, within a particular depth which depends on these factors. In this study two methods were used to incorporate the PCMs into concrete to form PCM/concrete composite panels. 1.

microencapsulated parafﬁn was added to fresh concrete during the mixing process.

butyl stearate was vacuum impregnated into lightweight aggregate which was then included in the concrete mix design.

The main objectives of this study were to: (i)

compare the effectiveness of the two methods of incorporating phase-change materials into concrete at increasing thermal storage capacity.

(ii) compare the effective depth of each PCM within the panel in order to inform future design of PCM-concrete panels so that the efﬁciency of the thermal storage behavior of the PCM/concrete composite material can be optimised.

(iii) To evaluate the effects that the phase-change materials have on the thermal conductivity strength, and density, of concrete.

2. Methodology To compare the two methods selected for combining the PCMs with concrete, six panels in total were manufactured, two panels of each type of PCM/concrete composite and two control panels with no PCM. In the following sections of this paper the panels containing the lightweight aggregate impregnated with butyl stearate are referred to as LWA/PCM and the panels containing microencapsulated paraffin are referred to as ME PCM. Each concrete panel was 200mm x 200mm x 200mm. A panel depth of 200mm was chosen to reflect the typical thickness of an internal wall within a building. In order to record the internal temperatures within the panels during testing three thermocouples were cast into each panel at equal depth intervals of 50mm. Thermocouples were also placed on the front and rear faces. After casting, the concrete panels were cured for 28 days in accordance with IS EN 123902. As moisture content has a significant influence on the thermal conductivity of materials the panels were allowed to dry out for 28 days after curing. The moisture content for all panels was less than 4% prior to measuring the thermal conductivity of the panels. To be able to accurately compare the heat transfer behaviour of the six panels, and ensure that any differences in thermal mass behaviour are due solely to the parameters of the panels, it is required that each panel is exposed to the same amount of thermal energy using the same mechanism of heat transfer. In order to replicate thermal energy transfer from the sun while controlling the amount of thermal energy that each panel is exposed to, radiation was selected as the mechanism of heat transfer. A particular artificial light source (Follow 1200 Pro lamp) was used with which it is possible to control the wavelength and intensity of the electromagnetic waves that are emitted. The lamp produces light across the three wavelengths of light – visible, infrared and ultraviolet – and mimics daylight. In order to exclude the environmental effects such as temperature variation in the test room, an insulated and airtight light box was designed and constructed as shown in Figure 2. Air temperature within the light box was recorded during all tests and the temperature difference between the internal air and the panel was monitored. Convective heat transfer from the internal air to the panel was considered to be minimal in comparison to the radiative heat from the lamp and so was omitted from consideration. To establish the optimum size for the light box, initial tests were carried out with the lamp to determine the light intensity (Lux) and spread of light that reaches a surface positioned at particular distances from the lamp. The heat energy reaching the surface was also measured in these tests using a pyronometer. The dimensions of the light box were optimised to ensure that the thermal energy from the lamp was evenly spread across the surface of the panel and that the intensity of energy was sufficient to heat the full depth of the panel within 12 hours. In order to ensure that an equal amount of additional latent heat capacity was added to each type of panel, the actual latent heat

Niall, West McCormack Paper.indd 4

14/11/2016 10:01

Assessment of two methods of enhancing thermal mass performance of concrete through the incorporation of phase-change materials

1400mm

100mm thick rigid insulation

1000mm

400 mm

Follow 1200 Pro Lamp

200 mm

Thermocouples

Data logger

Figure 2. Schematic of the light box design

capacity and melt temperatures of the PCMs was determined by Differential Scanning Calorimetry (DSC) tests. A scanning rate of 5oC/ min was used across a temperature range of -30oC to 50oC. The tests were repeated three times on each sample. A summary of the results is shown in Figure 3. It was found that the latent heat capacities were 97J/g and 172J/g for the microencapsulated parafﬁn and butyl stearate respectively.

The lightweight aggregate/PCM, composite was manufactured by the author in the laboratory. Preliminary tests were carried out on three types of lightweight aggregate to establish which type had the greatest absorption capacity – an expanded clay called LECA, an expanded ﬂy ash called LYTAG, and pumice. The samples were dried in an oven. Then half of each sample was weighed and immersed in water for 24 hours. The other half of each sample was weighed and placed in a dessicator. Water was poured over the sample and a vacuum was applied for 40 minutes. The samples were subsequently surface-dried and weighed. The mass and volume of water absorbed was determined and it was established that LECA possessed the highest absorption capacity. The LWA/PCM composite was made by vacuuming the exact required quantity of butyl stearate into the LECA (Figure 5).

Heat flow Vs Temperature for all PCMs - Heating Micronal 40

Butyl stearate

Figure 5. Manufacture of the aggregate/PCM composite Peak at 16.9oC

Heat flow (mW)

25 Peak at 20.09oC

-40

-30

-20

-10

0 10 Temperature (oC)

Figure 3. Heat ﬂow V’s temperature for PCM.

A microencapsulated paraffin product called Micronal was used in the microencapsulated PCM panels. Micronal is produced by BASF and comes in powder form, (Figure 4). Previous research studies, ( [7] and [9] ) concluded that 5% by mass of concrete is the optimum quantity of Micronal to be used in a concrete mix application. Higher quantities of Micronal yielded impractically low concrete strengths and also caused significant reduction in the thermal conductivity and density. This tended to counteract the increase in thermal storage capacity as heat flow into the panel was reduced. The latent heat capacity provided by incorporating 5% Micronal into the panels equated to 93120 joules.

Figure 4. 1.44kg of Micronal DS 5040X

Thermal conductivity is a key parameter in thermal mass behavior. In particular for this study once the heat is absorbed at the surface of the panel, the conductivity of the panel material will directly influence the rate of heat transfer, that is the heat flux, through the sample and hence the rate of phase change of the PCM within the panel. To determine the thermal conductivity of the panels a hot plate apparatus was used to create a steady state temperature differential across the samples, (Th – Tc) (figure 6). Tc was typically around 20oC and Th was typically around 40oC. The heat flux exiting the cooler surface was measured using a heat flux pad and the measurement is given in W/m2 which is equivalent to joules/(sec m2), that is q/At (Eq 7). The depth ‘d’ of the samples is known and hence the conductivity can be calculated from: k=

q d . (W/mK) (7) At (Th–Tc)

Figure 6. Concrete samples placed in the hot plate rig

Niall, West McCormack Paper.indd 5

14/11/2016 10:01

SDAR Journal 2016

The mass and density of each of the panels were also recorded. To observe and record the thermal storage behaviour of the panels, light box tests were carried out in which each panel was placed in the light box, one at a time, and heated by the lamp for 12 hours. The panel was then allowed to cool for 12 hours while remaining in the light box. The temperatures of the front and rear surfaces and at 50mm depth intervals within the concrete panel were recorded during the heating and cooling periods. The recorded temperature data, together with the measured densities and thermal conductivities, were used to determine the thermal properties of each panel and to compare the thermal storage behaviour of the panels.

The density of both types of PCM/concrete composites was lower than the control concrete due to the lower density of the PCM relative to the density of cement paste (Table 1). The conductivity and density of the materials inﬂuence the thermal mass behavior. However, the effect that they have varies depending on the ratio of conductivity to density of the material (Ref Eq 3). Table 1. Conductivity and density of panels Panel Type

Density (kg/m3)

Conductivity (W/mK)

Control

2284

1.56

Control

2295

2.10

ME PCM 1

2075

1.20

3.1 The effect of PCMs on the properties of concrete

ME PCM 2

2112

0.98

The strength of the concrete mixes was determined by standard cube tests in accordance with I.S. EN 12390-3. The addition of both the microencapsulated PCM (ME PCM) and the LWA/PCM composite had an adverse effect on the strength of the concrete panels. Both types of PCM panels only achieved strengths in the order of 25MPa after 28 days (Figure 7). The loss of strength may be due to leaked PCM, or possibly as a result of damaged capsules, interfering with the hydration process and/or adversely affecting the bond between the cement paste and the aggregate.

LWA/PCM 1

2076

0.82

LWA/PCM 2

2010

1.18

3. Results and discussion

28 day strength Mpa 56 day strength Mpa 60 50

3.2 Thermal storage behaviour The speciﬁc heat capacity of a material is given by: Cp =

6Q (J/kgK) (8) m6T

where: 6Q = quantity of thermal energy transferred to material, (Joules) 6T = change in temperature of the material (oC). However, for a PCM/concrete composite material the heat capacity varies during the phase transition and therefore as proposed by [17], eq. (8) must be modiﬁed to include the temperature gradient over time:

40 30

A.q

Cp = mdt (J/kgK) (9) di

20 10 0 Control

ME PCM

LWA PCM

Figure 7. Concrete strengths achieved

Notwithstanding this the strengths achieved are still suitable for some structural applications, such as non-loadbearing facade panels and low-rise construction/domestic construction. The addition of both types of PCM resulted in a reduction of 40-45% in thermal conductivity of the concrete (Table 1). This is caused by the relatively low conductivity of the PCM material. A reduced conductivity is not necessarily a problem as the desired conductivity depends on the required time frame within which the phase-change must occur – 12 hours in this study. Notwithstanding this, it is important that the conductivity of the PCM/concrete composite is sufﬁcient to achieve a thermal diffusivity that will allow the heat to penetrate the full depth of the panel and melt all of the PCM so that the latent heat capacity provided by the PCM is fully utilised.

where ‘A’ is the area of the sample (m2), q is the thermal energy supplied to the sample (W/m2), m is the mass (kg), dT/dt = increase in sample temperature in a given time step (oC/s). In the light box tests carried out as part of this research each of the panels was exposed to equal amounts of thermal energy from the lamp over an equal time period of 12 hours hence the ‘q’ value is the same for each panel. Also the area exposed to the light is the same for each panel at 0.2m2. Hence the overall thermal storage capacity of the panels can be compared by evaluating the mass x dT/dt value for each panel. The heat flux, that is the rate of heat transfer through the PCM/ concrete composite material, will vary throughout the depth of the material as the PCM changes phase. As a result, the heat flux transferred to the surface of the sample is overestimated with respect to the internal temperature gradient over time which leads to an overestimate of the overall thermal storage capacity. To overcome this issue the applied heat flux ‘q’ is left in the equation as a constant and only the data from the three internal thermocouples at 50mm, 100mm and 150mm is considered.

Niall, West McCormack Paper.indd 6

14/11/2016 10:01

Assessment of two methods of enhancing thermal mass performance of concrete through the incorporation of phase-change materials

Table 3. Additional thermal storage provided by PCM panels measured at 100mm depth

Overall thermal storage recorded at 50mm Poly. (Control C3)

Cp/(Aq)

225 215 205 195 185 175 165 155 145 135 125 115 105 95 85 75 65 55 45 35 25 00:00

ME PCM

LWA PCM

ΔT in Panel(oC)

% Overall thermal storage relative to control panel

Control

100.0

ME PCM

147.0

LWA PCM

143.0

Panel Type

01:00

02:00

03:00

04:00

05:00

06:00

07:00

08:00

09:00

10:00

11:00

12:00

Time (hrs)

Figure 8. Curves showing relative overall thermal capacity at 50mm

The temperature data for each panel was analysed and the time taken for each 1oC increase in temperature throughout the 12 hour period was determined, ie dT/dt over time. Each dT/dt value is then multiplied by the mass of the relevant panel and the reciprocal of the result is calculated, ie, 1/(m(dT/dt). This value is then plotted against time to observe how it varies over the 12 hour heating period. The higher the value of 1/(m(dT/dt) the higher the thermal storage capacity. The overall area under the resulting curves is indicative of the overall thermal capacity and a comparison of the thermal storage capacity of the panels was made Figure 8 shows a plot of the relative overall thermal storage, as recorded at 50mm depth throughout the 12-hour heating period. It is clear that the panels containing PCM provide greater thermal storage capacity. As conﬁrmed by computing the area under each of the curves, the lightweight aggregate panels (LWA/PCM) provide the highest additional overall thermal capacity measured at a depth of 50mm. Table 2. Additional thermal storage provided by PCM panels measured at 50mm depth Panel Type

ΔT in Panel(oC)

% Overall thermal storage relative to control panel

Control

100.0

ME PCM

157.5

LWA PCM

161.7

The percentage of additional thermal storage and thermal mass provided by the PCM panels was determined and the results are shown in Table 2. It is noted that the LWA/PCM panel provides the largest increase in thermal storage of 61.7%. The panel with microencapsulated PCM (ME PCM) also provides a signiﬁcant increase in thermal storage of 57.5%. Table 3 and Table 4 show the equivalent results computed from the data recorded at 100mm depth and 150mm depth.

Table 4. Additional thermal storage provided by PCM panels measured at 150mm depth Panel Type

ΔT in Panel(oC)

% Overall thermal storage relative to control panel

Control

100.0

ME PCM

152.0

LWA PCM

147.0

At each thermocouple location the LWA/PCM panel displays the lowest change in temperature over the 12-hour period. It can be noted that the overall thermal storage of the PCM panels reduces with depth relative to the control panel. This behavior is due to the lower conductivity and higher thermal storage capacity of the PCM panels which resulted in reduced thermal diffusivity and, in turn, reduced the effectiveness of the PCM as depth increased as the heat took longer to reach the PCM. As shown in Figure 9, the LWA/PCM panels displayed the lowest diffusivity. This means that the heat took longer to penetrate 100mm in the LWA/PCM panels, so over the 12hour period the overall heat reaching 100mm depth in the LWA/PCM panels is less than that in the control panel and also the ME PCM panels. Hence the PCM becomes less effective with increasing depth. In a real application the level of exposure to solar energy depends on both local climate and the position of the concrete element within the building, ie, exposure to daylight. Although not considered in this research, other sources of heat within a building arise from occupancy, equipment and lighting that depends on building use. Also, the build-up of heat within a building will depend on the level of insulation provided and the airtightness of the construction. Ultimately, in a real life scenario the effective depth of the PCM will depend on all these variables. Building energy performance software would be required to analyse the thermal behavior of a particular building taking into account all the relevant variables. This study has shown that in applications where the heat energy is reaching up to a depth of 100mm into the composite PCM/concrete walls, the LWA/PCM composite is more efﬁcient at providing greater thermal storage capacity.

Niall, West McCormack Paper.indd 7

14/11/2016 10:01

SDAR Journal 2016

Control C3 Log. (Control C3)

α/Aq x 10-6

24 23 22 21 20 19 18 17 16 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0 00:00

01:00

02:00

03:00

ME PCM 1 Log. (ME PCM 1)

04:00

05:00

06:00

07:00

It is also observed that the rate of decrease in temperature is similar for the control panel and the LWA/PCM panel with the LWA/PCM panel showing a slightly higher rate of heat loss at the surface. Calculation showed that outside of the phase-change period, the control panel and the LWA/PCM panel have a similar heat storage capacity. The higher conductivity and density of the control panel is contributing to the slightly higher thermal inertia of the control panel.

LWA1 Log. (LWA1)

08:00

09:00

10:00

11:00

12:00

Time (hrs)

Figure 9. Relative thermal diffusivity recorded at 50mm

Figure 10 shows the relative thermal inertia recorded at a depth of 50mm. It is noted that despite having the lowest thermal diffusivity, the LWA/PCM panel displays the lowest thermal inertia. This is caused by the low conductivity and density of the LWA/PCM panels. Control

Poly. (ME PCM 1)

Poly. (LWA PCM 1)

850 800

750

I = √((

)

700 650 600 550 500 450 400 350 300

250 00:00

01:00

02:00

03:00

04:00

05:00

06:00

07:00

08:00

09:00

10:00

11:00

12:00

Time (hrs)

Figure 10. Relative thermal inertia recorded at 50mm

3.3 Cooling behavior After the 12-hour heating period the Follow Pro 1200 lamp was switched off and the panels remained in the light box for a further 12 hours to cool down naturally while the temperature data was recorded. The panels did not cool down sufﬁciently within this time period to induce a solidiﬁcation phase-change of the PCM within the panels. However, some observations and comparisons can be made regarding the rate of cooling of the front surface of the panels. In order to induce a solidifying phase-change the panels were also cooled in a fridge. In these tests all faces of the panel are exposed and the rate of temperature decrease is high. An unexpected observation from the natural cooling data is that the front face of the ME PCM panel cooled at a higher rate than the control panel, despite having a lower conductivity and lower density. However, calculations showed that the ME PCM material has a lower overall thermal capacity Cp than the control panel outside of the phase-change period. As there was no phase-change taking place during this cooling period, it can be assumed that the Cp value for the ME PCM panel is lower than Cp for the control. This leads to a higher thermal diffusivity in the ME PCM panel, facilitating the release of heat from the front of the panel.

A notable observation from the fridge cooling tests is that the LWA/ PCM panel loses heat at a slightly slower rate, displaying a higher thermal inertia. The thermal capacity of the concrete matrix in the LWA/PCM panel is the same as the thermal storage capacity of the control panel. However, the heat released by the solidifying PCM slows down the cooling of the LWA/PCM panel. It was generally noted that the ME PCM and LWA/PCM panels have similar cooling rates and both panels display thermal inertia relative to the control panel. This is due to the release of heat from the PCM as it solidifies which slows down the rate of cooling. The study of the data collected during the natural cooling of the panels within the light box highlighted a critical issue with the use of PCM/concrete composites in buildings, which is that the indoor temperature must fluctuate above the melting temperature and below the freezing temperature of the PCM within a 24-hour period. If this range of temperature fluctuation does not occur then the PCM will not discharge latent heat energy and will not have the capacity to absorb more heat the following day. The fluctuation in the indoor temperature depends on both the local climate and the level of insulation in a building. Modern buildings tend to be highly-insulated to prevent loss of heat energy though high levels of insulation may hinder the performance of a PCM thermal energy storage element within a building. For this reason the use of PCM/concrete composites for enhanced thermal mass behaviour is more suitable for the refurbishment of buildings that have poor levels of insulation and airtightness. For example, in an analysis carried out by the World Business Council for Sustainable Development [18], it was stated that there is currently a stock of more than 80 million buildings in Europe that were constructed between 1950 and 1975 in an era when energy performance was not included in the design criteria. Rather than demolish these buildings it may be argued that a more sustainable solution would be to refurbish the façade of the building using panels with enhanced thermal mass properties that would improve the energy efficiency of these buildings.

4. Conclusions Based on the results of the analysis presented in this paper the following conclusions can be made: • Up to a depth of 100mm the concrete panels containing the LWA/PCM composite provided a greater increase in thermal storage capacity compared to the control panels and ME PCM panels; • The LWA/PCM panel displayed the lowest increase in temperature throughout the 12-hour heating period;

Niall, West McCormack Paper.indd 8

14/11/2016 10:01

Assessment of two methods of enhancing thermal mass performance of concrete through the incorporation of phase-change materials

• The addition of both types of PCM caused a reduction in thermal conductivity and density. This resulted in lower thermal diffusivity in the panels containing PCM; • As depth increases the level of thermal storage provided by the ME PCM panel approaches the storage provided by the LWA/PCM panel and, at a depth of 100mm, the storage provided by the ME PCM panel was slightly greater than the LWA/PCM panel. Hence if the local conditions allow the heat energy to penetrate deeper than 100mm, the ME PCM composite material will provide a greater increase in thermal storage capacity; • The effectiveness of both types of PCM in increasing the overall thermal storage of the concrete panels relative to the control panel reduces with depth. This is due to the fact that the thermal diffusivity of the PCM panels is lower than the control panels, hence the heat takes longer to penetrate to a depth of 100mm in the LWA/PCM and ME PCM panels; • As thermal diffusivity is the parameter that is hindering the effectiveness of the LWA/PCM composite, improving the conductivity of the LWA/PCM panels would further enhance the thermal performance of the material; • The design of PCM/concrete composite applications in buildings requires careful consideration of local climate, building use and construction details. For the PCM to be effective it is critical that the temperature in the area that the PCM is located varies above and below the melt temperature range of the selected PCM within a diurnal period.

5. Further research Further research is currently being carried out to investigate methods of improving the thermal conductivity of concrete containing lightweight aggregate/PCM composite. It is also planned to construct a number of full-scale huts using cladding panels containing an inner leaf wall constructed with PCM/concrete composite. These huts will be used to record thermal data to enable an assessment of thermal mass behaviour of the PCM/concrete inner wall under real conditions.

References [1]

de Saulles, T., 2012. Thermal Mass Explained. The Concrete Centre

[2]

Sharma, A., Tyagi, V., Chen, C. and Buddhi, D., 2009. Review on thermal energy storage with phase change materials and applications. Renewable and Sustainable Energy Reviews, 13(2), 318–345.

[3]

Pomianowski, M., Heiselberg, P. and Jensen, R.L., 2012. Dynamic heat storage and cooling capacity of a concrete deck with PCM and thermally activated building system. Energy and Buildings, 53, 96–107

[4]

Kuznik, F., David, D., Johannes, K. and Roux, J., 2011. A review on phase change materials integrated in building walls. Renewable and Sustainable Energy Reviews, 15(1), 379–391.

[5]

Kosny, J., and Kossecka, E., 2013. Understanding a Potential for Application of Phase-Change Materials (PCMs) in Building Envelopes. ASHRAE Transactions 119(1), 1-11

[6]

Pasupathy, A., Velraj, R. and Seeniraj, R.V., 2008. Phase change material-based building architecture for thermal management in residential and commercial establishments. Renewable and Sustainable Energy Reviews, 12(1), 39–64

[7]

Hunger, M., Entrop, A. G., Mandilaras, I., Brouwers H. J. H. and Founti, M., 2009. The behaviour of self-compacting concrete containing microencapsulated Phase Change Materials. Cement and Concrete Composites, 31(10), 731-743

[8]

Eddhahak-Ouni, A., Drissi, S., Colin, J., Neji, J. and Care, S., 2014. Experimental and multi-scale analysis of the thermal properties of portland cement concretes embedded with microencapsulated Phase Change Materials (PCMs). Applied Thermal Engineering, 64(1-2), 32-39

[9]

Fenollera, M., Miguez, J. L., Goicoechea, I., Lorenzo, J. and Alvarez, M. A., 2013. The Inﬂuence of Phase Change Materials on the Properties of Self-Compacting Concrete. Materials, 6(8), 3530-3546

[10] Baetens, R., Jelle, B.P. & Gustavsen, A., 2010. Phase change materials for building applications: A state-of-the-art review. Energy and Buildings, 42(9), 1361–1368. [11] Cabeza, L.F., Castellon, C., Nogues, M., Medrano, M., Leppers, R. & Zubillaga, O., 2007. Use of microencapsulated PCM in concrete walls for energy savings. Energy and Buildings, 39(2), 113–119. [12] Kuznik, F., Virgone, J., 2009. Experimental assessment of a phase change material for wall building use. Applied Energy, 86(10), 20382046. [13] Ling, T.-C. and Poon, C.-S., 2013. Use of phase change materials for thermal energy storage in concrete: An overview. Construction and

Building Materials, 46, 55–62 [14] Lee, T., Hawes, D. W., Banu, D. and Feldman, D., 2000. Control aspects of latent heat storage and recovery in concrete. Solar Energy Materials and Solar Cells, 62(3), 217–237 [15] Zhang, D., Li, Z., Zhou, J. and Wu, K., 2004. Development of thermal energy storage concrete. Cement and Concrete Research, 34(6), 927– 934 [16] Waqas, A., Ud Din, Z., 2012. Phase change material (PCM) storage for free cooling of buildings – A review. Renewable and Sustainable Energy Reviews, 18, 607-625. [17] Pomianowski, M., Heiselberg, P., Jensen, R.L., Cheng, R. and Zhang, Y., 2014. A new experimental method to determine speciﬁc heat capacity of inhomogeneous concrete material with incorporated microencapsulatedPCM. Cement and Concrete Research, 55, 22–34 [18] World Business Council for Sustainable Development, (2009). Energy Efﬁciency in Buildings –- Transforming the Market. Available at http://www. wbcsd.org/transformingthemarketeeb.aspx [Accessed 15th May 2016]

Niall, West McCormack Paper.indd 9

14/11/2016 10:01

School Elec Eng 2016:Layout 1

10/11/2016

09:04

Page 1

Institiúd Teicneolaíochta Átha Cliath Dublin Institute of Technology

School of Electrical and Electronic Engineering The School of Electrical and Electronic Engineering, Dublin Institute of Technology (SEEE), 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 across a variety of full-time and parttime options. The School also focuses on applied research with a strong emphasis on producing useful and novel ideas to help Irish industry compete globally. 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.

SEEE Programmes Level 9 (Masters) MSc in Energy Management

DT711 or DT015

ME in Sustainable Electrical Energy Systems

DT704 or DT705

MSc in Electronic and Communications Engineering

DT085 or DT086

Level 8 (Hons) BE in Electrical and Electronic Engineering

DT021

BE in Computer and Communications Engineering

DT081

BSc in Electrical Services and Energy Management

DT035, DT712 or DT018

BSc in Networking Applications and Services

DT080B

Level 7 BEngTech in Electronic and Communications Electrical Services Engineering Engineering

DT008

BEngTech in Electrical and Control Engineering

DT009

Level 7 in Electrical Services Engineering BEngTech BE in Networking Technologies BTech

DT010

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 4617/4650/4575 Email: seee.admin@dit.ie www.seee.dit.ie

DT080A

Irish Large Scale Solar PV Opportunities: A viability analysis SULRULWLVLQJ WKH LQĂ XHQFH of System Harmonics

Andrew Hogan

PREMIUM POWER, IRELAND ahogan@premiumpower.ie

Keith Sunderland

DUBLIN INSTITUTE OF TECHNOLOGY, IRELAND

Hogan Sunderland Paper.indd 1

14/11/2016 09:14

SDAR Journal 2016

Abstract

1. Introduction

This paper considers the techno-logistical considerations

IIn the context of a sustainable energy future for Ireland, the Irish Government has implemented policy aimed at reducing carbon emissions, reliance on fossil fuels and providing the means to accelerate growth within the renewable energy sector. These policies promote a sustainable renewable energy sector in Ireland. The Government’s National Renewable Energy Action Plan (NREAP) for instance, sets a mandatory target of 16 % for renewable energy in three key areas — electricity, heat and transport. The designated contributions towards each area include 40% for electricity (RES-E), 12% for heat (RES-H) and 10% for transport (RES-T). Such policies, however, have the potential to introduce major technical problems for the transmission/distribution system operators (TSO/DSO) in terms of availability, power system stability and power quality (PQ). These technical problems are already prevalent and have led to a cap of 50% on synchronous non-synchronous penetration (SNSP) on national grids (Eirgrid, 2009).

involved in designing for a large scale (>100 kW) distributed generation (DG) opportunity. The logistical considerations involve site feasibility assessment in terms of resource, land requirements and PV plant design, whereas the technical considerations prioritise the impact that PV inverters can have on the performance of a power system; particularly in the context of the detrimental effects manifested by harmonic distortions. The analysis demonstrates that without proper consideration of the PV system conﬁguration and how PV inverters are employed, capacity rating breaches will affect both the DSO and the consumer. Ultimately, the results point towards a need for policy measures that encourage power system studies at the logistical stage of development to encourage what is ‘the ultimate’ in a sustainable energy resource. Keywords: Solar photovoltaic (PV), feasibility study, PV system design, grid codes, harmonic distortion. Glossary AC

Alternating Current

PCC

Point of Common Coupling

DG DSO

Distributed Generation Distribution System operator

PQ Power Quality RES-E Renewable Energy SupplyElectricity

EPIA

European Photovoltaic Industry Association

RES-H Renewable Energy SupplyHeat

IWEA Irish Wind Energy Association

RES-T Renewable Energy SupplyTransport

Low Voltage

SNSP Synchronous NonSynchronous Penetration

Medium Voltage

TSO

Photo-Voltaic

VTHD Voltage Total Harmonic Distortion

PFC

Power Factor Correction

PCC

NREAP National Renewable Energy Action Plan

Transmission System Operator

Point of Common Coupling

In Ireland a large emphasis has been placed on wind. Wind, particularly large-scale wind, is generally perceived as the primary renewable energy technology that will help Ireland meet its 2020 target of 40% of all electricity to come from renewable sources. In 2016, the Irish Wind Energy Association quoted 2,441 MW of installed wind capacity in the Republic of Ireland (IWEA, 2016). This is substantial when compared to the country’s actual electricity demand of 5000 MW. These ﬁgures would suggest that Ireland is not too far away from achieving the 40% RES-E target. However, antagonistic factors including wind resource intermittency and negative public perception of large wind plant, as well as the environmental challenges such deployments face, suggest that wind alone may not be enough to meet the 40% RES-E target. Solar PV, which is often overlooked and underestimated in Ireland, is a renewable technology that could be used to complement wind power and add to the current renewable power mix in Ireland. The notion that Ireland does not possess a solar resource capable of supporting a sustainable PV sector has been proven incorrect in recent times. The European Photovoltaic Industry Association’s (EPIA) global outlook for PV in 2014 showed how Ireland’s solar resource is equivalent to that in the UK and parts of mainland Europe (EPIA, 2014). Solar Power Europe reported that by the end of 2015 the UK had an installed PV capacity of 9.2GW (SPE, 2015), which is extraordinary when compared to Ireland’s estimated 3MW (TEA, 2014). According to the Irish Solar Energy Association however, Ireland’s solar insolation levels are 78% of the level received in Madrid and are equal to the levels experienced across sites in the UK (ISEA, 2014) The statistics presented in the previous paragraph suggest that the solar resource should not be considered as the exclusive barrier to PV deployment in Ireland. One can therefore extrapolate that the challenges facing the PV sector are actually more related to development opportunities and the considerations therein. This paper will outline a structured approach to developing a PV system in Ireland. Starting with the high level design of a 1 MW PV facility, the paper will describe the key steps involved in a PV

Hogan Sunderland Paper.indd 2

08/11/2016 14:33

Irish Large Scale Solar PV Opportunities: A viability analysis prioritising the inﬂuence of System Harmonics

feasibility study before carrying out harmonic analysis on the proposed 1MW PV system installation. In this regard, the advanced power systems modelling tool (DIgSILENT) is employed to assess a 1MW PV facility that is connected to the distribution network at medium voltage (MV) i.e. 10kV – 38kV. The results of the harmonic emissions investigation are subsequently used to determine if a harmonic ﬁltration unit is required. Harmonic ﬁlters can be designed to limit the impact harmonic distortion levels can have on electrical power systems.

2. Background Resource estimates prove that largescale solar PV can be a viable alternative source of energy (ISEA, 2014). Such a resource could be used to complement the large volumes of installed wind capacity in Ireland. However, evidence of the resource alone will not be enough to achieve a sustainable PV sector in Ireland. Feasibility studies that not only look at resource but focus on strategies that help improve conversion efficiencies, from energy production to final consumption, are essential to accomplish a successful/attractive PV sector. Selecting a site close to where the load is consumed immediately improves system performance and contributes positively to the overall power demand of the network. Research identifies transmission and distribution network losses as the single largest use of electrical energy in any power systems (Targosz, 2008). There are also underlying factors that are often neglected, but are crucial to sustained PV development and growth. The DSO in Ireland, through industry standards such as EN50160, must ensure that Voltage Total Harmonic Distortion levels (VTHD) do not exceed 8% for 95% of the time. Through the Irish distribution codes, the DNO has designated an upper limit of 2% VTHD on harmonic emissions for all connected generators. The attention focused on non-linear PV inverters in this paper, prompted by research carried out by Benhabib (2007), relates to the technical issues that can arise where multiple sources of harmonic distortion are connected to any given power system. Harmonics, relating to solar PV, and as a power quality (PQ) concern for power network operators, are of particular interest for two reasons. Firstly, solar PV systems, by virtue of how they generate electricity, utilise non-linear high-frequency switching devices (inverters) to produce AC power, which is required for synchronisation with electricity grids. Secondly, Benhabib’s (2007) research, which involved 96 residential homes where small grid-tied PV systems were connected to the DSO’s 400V network, demonstrated how connecting an accumulative non-linear load (inverter) in parallel to a common electrical bus systematically caused the harmonic distortion levels to rise. The final scenario associated with Benhabib’s work assessed 96 gridtied inverters coupled with four non-linear loads and the associated analysis concluded that the VTHD levels could reach 51% (Benhabib, 2007). High levels of VTHD on the Irish distribution network will lead to PV system curtailments. Such action is unavoidable if compliance with the EN50160 standard cannot be achieved, which will ultimately lead

to the failure of critical network components. These failures include the overheating of transformers, overloading of neutral conductors, nuisance tripping of circuit breakers, over-stressing of PFC (power factor correction) capacitors, overheating of windings in induction motors and likely damage to converters, telecommunications and other electronic equipment.

3. Site feasibility study Large-scale renewable energy projects require careful consideration of critical criteria in order to achieve optimum performance. Such criteria includes resource, land requirements, customised system design and proximity to distribution load centres, each one interlinked and interdependent. The methodology involved in allocating a site to install the PV generator is illustrated in Figure 1. A consideration of the site requirements as they apply to a 1MW PV facility is considered in the following sections. This includes a system conﬁguration (Section 3.1) and land availability (Section 3.2) perspective, with Sections 3.3 and 3.4 providing a resource and demand consideration respectively.

Figure 1: Feasibility study and PV system design process chart

3.1 PV System Design The number of Mitsubishi MLE-260 PV modules per string, and subsequently the number of PV strings within an array, is governed by the PV inverters DC electrical input characteristics. The selected PV inverter, which is an SMA Sunny T60, is manufactured to EN 50438 and tested to harmonic emissions standard EN 61000-3-12. The voltage, within a series-connected string, increases by 30.9V as each Mitsubishi 260 W PV module is added. In this situation the current remains constant at 8.59A. To achieve maximum power at the PV inverters input terminals, additional strings are connected in parallel. Connecting similar-sized PV strings in parallel increases the derived current by 8.59A with every string added in parallel. In this regard, adding panels in parallel has no effect on the system voltage. So, for 1kV (constant) at maximum power: •

No. of inverters: total system capacity / inverter max AC power output rating, (1000kW/60 kW) = 17 inverters

•

No. of modules per string: (maximum inverter input voltage/PV module output voltage), 1000 V/30.9V = 32 PV panels/string

Hogan Sunderland Paper.indd 3

08/11/2016 14:33

SDAR Journal 2016

•

No. of strings/array: (maximum inverter input current/PV module output current), 110 A / 8.59 A = 13 strings/array

A summary of the system conﬁguration is provided in Figure 2.

The land requirements calculated above are consistent with industry guidelines. Miller (2012), who is supported by Brooks (2013), explains how the PV array area for a well-designed 1MW facility can vary between 10,000 m² (2.5 acres) to 20,000 m² (five acres) (Miller, 2012). The variance in land area is dependent on solar resource and V-module efficiencies. 3.3 Resource and Energy Yield Analysis Solar irradiance maps, as illustrated in Figure 3, are employable to identify locations in Ireland with a suitable solar resource for largescale power generation. In the analysis presented here, southern Ireland is prioritised, with consideration specifically applied to Valentia, Bantry, Roches Point, Youghal and Rosslare.

Figure 2: Conﬁguration of 1MW PV array.

Once a range of locations with appropriate irradiance levels is identiﬁed, historical daily resource levels are obtainable through proprietary software such as RETScreen (NRC, 2015). Monthly energy yield estimates are subsequently determined for each location as illustrated in Figure 4, which compares the potential energy yield at each location considered.

3.2 Land Requirements Previous research carried out by SRA (2012) demonstrates how ground-mounted PV systems are land intensive with respect to the power they produce (SRA, 2012). The total land requirement for the 1MW facility has been estimated using the dimensions listed in Figure 2 as follows: •

Total area of PV array: (length of string) x (width of 13 strings + associated row spacing), 43 m x (13 x 2) = 1,118 m²

•

Total Land Requirement: (area of 1 array) x (total no. of arrays) 1,118 m² x 17 = 19,006m² or 4.7 acres

Figure 4: Monthly energy yield comparative analysis.

3.4 Load Distribution Centre To maximise the potential yield of a PV system, system losses should be minimised by ensuring the solar power is consumed as close as possible to where it is produced. Therefore, reducing the effects of energy distribution systems (such as cable systems) is a priority in this regard. In the context of the counties considered, Co Cork, as portrayed in Table 1, has the largest electrical load requirement.

Table 1: Electrical Energy Balance for Southern Regions (AIRO, 2014)

2014 Electricity Usage

National National Cork Wexford Kerry ktoe GWh GWh GWh GWh

Industry, Commercial & Public Services Electricity

1434

16674

3815

504

518

Residential Electricity

663

7709

1781

245

248

Residential & Industrial Electricity Consumption

2097

24384

5596

749

766

Figure 3: Solar radiation map.

Hogan Sunderland Paper.indd 4

08/11/2016 14:33

Irish Large Scale Solar PV Opportunities: A viability analysis prioritising the inﬂuence of System Harmonics

Therefore, from an installation location perspective, the priority locations are Bantry, Roche’s Point or Youghal, all located in County Cork.

4. Power system modelling The proposed 1MW PV facility has been simulated using the system design in Section 3. The harmonic analysis methodology was conducted under the following test conditions: •

Obtain source impedance data from ESB Networks for two substation classiﬁcations; see scenario one (high impedance) and scenario two (low impedance) below.

•

Model 17 x PV inverters in DigSILENT software using harmonic data from SMA Sunny T60 EN50438/2013 test certiﬁcates (SMA, 2015).

•

17 x 60 kW inverters are connected in stages with the aggregated voltage harmonic emission levels measured at each stage.

•

Scenario analysis to assess the impact of source impedance and cable capacitance using parameters in Table 2. Table No.2: Parameters for Harmonics Analysis Parameter

Scenario 1

Scenario 2

PV Installed Capacity

1 MW

LV Transformer Size

2 MW

MV Substation Voltage

10 kV

20 kV

MV Substation Impedance

4.8 1

2.8 1

MV Cable Size

120 mm²

MV Cable Distance (short)

50 meters

MV Cable Distance (actual)

2 kM

•

Record % VTHD values at PV system LV bus and MV substation (PCC).

•

Harmonic emission levels are compared to % VTHD limits set out in the distribution code for generators.

Figure 5: Voltage harmonic analysis – 50m cable run (high Z network).

4.1 Harmonic Analysis Results The harmonic analysis results, outlined in Figure 5, illustrate clearly that the harmonic emission levels increase proportionally with the addition of PV inverters. The 17 x 60kW PV inverters generate a current that is disproportional to the utility supply, thereby distorting the network voltage. Non-linear loads, such as PV inverters, draw a non-linear current that is disproportional to the applied voltage (ABS, 2006). Network/source impedance (Z), obtained from the local utility company, plays a vital role in determining the magnitude of voltage harmonic distortion, which is illustrated in Figures 6 and 7. This is because the voltage harmonic distortion is the manifestation of non-linear current interactions with the networks impedance. Another significant factor, realised while modelling the MV cable, was the impact that cable capacitance has on the overall VTHD levels. The harmonic emission levels observed, when a 120 mm² 50 m MV cable run was simulated, were almost 20% - 40% less than those calculated with the actual cable length of 2kM (see Figure 5 and 6 and Table 3). All cables contain resistive, inductive and capacitive characteristics. The capacitive element is dominant in high and medium voltage cables. In relatively long high/medium voltage cable runs, the cable capacitance amplifies the network harmonics, which is evident in the results recorded in Figure 6 and Table 3. The cells highlighted in red within Table 3 represent the points where the voltage harmonic distortion levels breach the distribution codes limit of 2% for grid-connected generators. Although breaches occur when a cable run of 50m is used, the results reveal how an influx

Figure 6: Voltage harmonic analysis – 2kM cable run (high Z network).

Figure 7: Voltage harmonic analysis – 2kM cable run (low Z network).

Hogan Sunderland Paper.indd 5

08/11/2016 14:33

SDAR Journal 2016

of long cable runs and non-linear equipment can exacerbate the magnitude of the distribution systems voltage distortion. 4.2 Mitigation To mitigate against the high harmonic currents, a passive harmonic filtering solution is sized using DigSILENT software. Passive filters are L-C circuits where an inductor (L) is connected in series with a capacitor (C). The passive filtering solution, which is connected in parallel with the non-linear PV inverters, is used to absorb the current harmonics; reducing the overall VTHD (see comparative results in Table 4). With all PV inverters operating at 100% (full) capacity, the filters reactive power and tuning frequencies were adjusted and the VTHD levels at 10kV substation were observed. The filter set-up was continually adjusted until the VTHD levels fell below 2% at the 10kV point of common coupling (PCC). Table 3: Harmonics Results Summary

Scenarios

50 M Cable/OH Line 2 km Cable/OH Line Inverter kVA 400 V Bus 10 kV Bus 400 V Bus 10 kV Bus Output %VTHD %VTHD %VTHD %VTHD

0 x Inverter

0 kVA

4 x Inverter

240 kVA

1.60

1.74

2.94

2.90

8 x Inverter

480 kVA

2.99

2.76

4.51

4.40

12 x Inverter

720 kVA

4.18

3.88

5.72

5.38

17 x Inverter

1,020 kVA

5.64

5.25

7.49

7.25

Table 4: % VTHD results with and without passive ﬁlter

Scenarios

With Filter Without Filter Inverter kVA 400 V Bus 10 kV Bus 400 V Bus 10 kV Bus Output %VTHD %VTHD %VTHD %VTHD

0 x Inverter

0 kVA

4 x Inverter

240 kVA

0.43

0.46

2.94

2.90

8 x Inverter

480 kVA

0.86

0.91

4.51

4.40

12 x Inverter

720 kVA

1.28

1.38

5.72

5.38

17 x Inverter

1020 kVA

1.81

1.95

7.49

7.25

distortion, at both LV and MV systems, as additional PV inverters are connected in parallel to the network. Although PV inverters introduce non-linear currents to power systems and act as a source for VTHD, their impact on the power systems overall VTHD is minimal when compared to other influential factors. The research here also identifies the adverse effect long cable runs can have on the networks overall VTHD emission levels as a consequence of the associated increased capacitance. This is signified in Table 3 and Figure 6 where the voltage harmonic distortion was amplified by almost 50% when the cable run was increased from 50m to 2km. The analysis, however, identifies source impedance as having the greatest effect on the overall harmonic distortion levels. Although the PV inverters act as a harmonic source, high impedance networks significantly amplify the distortion levels. This amplification not only occurs at the point of common coupling (PCC) to the grid, but at the local switchgear where the PV inverters are connected as well. Simulations that consider the introduction of a passive filter, which can absorb harmonic currents, show how the VTHD on site can be significantly reduced (Table 4). Without the utilisation of a passive filter, it is unlikely that the PV system in Scenario 1 would ever be called upon to export power to the distribution network due to the high harmonic emissions caused by the PV inverters, cable capacitance and high source impedance.Compounded with this mitigation concern, other system performance viability impacts that are created from increased harmonic distortion proliferation include component failures, prolonged outages and reduced life cycles for PV plant. Such (harmonic) proliferations will effectively make PV in Ireland unsustainable and non-profitable without appropriate corrective measures. However, the research presented here suggests that the biggest challenge that may face PV developers in the future is not the problems caused by PV inverter induced harmonic emissions. Developers, who are granted a connection to the grid — but at a high impedance point — are inevitably at a disadvantage. They may incur higher installation costs, e.g. additional MV switchgear and filtration may be required to achieve compliance with distribution code limit of 2% VTHD at the PCC.

The cells highlighted in red within Table 3 represent the points where the voltage harmonic distortion levels breach the distribution codes limit of 2% for grid-connected generators. Although breaches occur when a cable run of 50m is used, the results reveal how an inﬂux of long cable runs and non-linear equipment can exacerbate the magnitude of the distribution systems voltage distortion.

5. Conclusion This paper presents a consideration of large-scale PV systems and the viability of same with a particular focus on the considerations/ ramiﬁcations of system harmonics for the distribution network operator. The results show a systematic rise in voltage harmonic

Hogan Sunderland Paper.indd 6

08/11/2016 14:33

Irish Large Scale Solar PV Opportunities: A viability analysis prioritising the inﬂuence of System Harmonics

References ABS. (2006). Control of Harmonics in Electrical Power Systems. New York: American Bureau of Shipping. AIRO. (2014). Census Mapping. Retrieved Nov 1st, 2015, from http://airo. maynoothuniversity.ie/mapping-resources/airo-census-mapping/local-authoritymodules Baggini, A. (2008). Handbook of Power Quality (1 ed.). Bergamo: John Wiley & Sons. Barrett, D. M. (2015). Transmission Networks – Electricity Markets. Dublin: DIT. Benhabib, M. (2007). Harmonic effects caused by large scale PV installations in LV network. Barcelona: Leonardo Power Quality Group. Brooks, W. (2013). PV Installation Professional Resource Guide. New York: NAPCEP. Chapman, D. (2001). Harmonics – Causes and Effects. London: Copper Development Association. DGS, G. E. (2009). Photovoltaic Systems: A Guide for installers, architects and engineers (2nd ed.). London: Earthscan. Dykes, P. (2011). EU Solar Days - Solar Energy in Ireland. Wexford: SEAI. ESBN. (2015). Distribution Code. Dublin: Distribution System Operator ESB Networks. Fassbinder, S. (2014). Passive Filters. European Copper Institute. Ging, J. (2013). Harmonic Issues and their Impact on Customer Connections. Dublin: EirGrid. ISEA. (2014). Submission for Green Paper on Energy Policy in Ireland. Dublin: Irish Solar Energy Association. IWEA. (2016). Wind Energy Statistics. Retrieved June 22nd, 2016, from http:// www.iwea.com/windstatistics Markvart, T. (2008). Solar Electricity (2ns ed.). West Sussex: John Wiley & Sons. Miller, A. (2012). Utility Scale Solar Power Plants: A Guide for Developers & Investors. International Finance Corporation (IFC). Montana, G. P. (Anon). The Power of Solar Energy. Retrieved November 7th, 2015, from http://www.montanagreenpower.com/solar/curriculum/lesson6.php Rashid, M. H. (2011). Electronics Handbook: Devices, Circuits and Applications (3rd ed.). Elsevier. SEAI. (2012). Best Practice Design Guide Photvoltaics. Dublin: Sustainable Energy Authority Ireland. Shah, N. (2013). Harmonics in Power Systems –- control, causes and effects. Alpharetta, GA: Siemens. SMA. (2016, April 22nd). Pricing for ST 60 PV Inverters. SMA. (2015). SUNNY TRIPOWER 60. Retrieved 11 11, 2015, from http://www. sma.de/en/products/solarinverters/sunny-tripower-60.html SPE. (2015). Global Market Outlook. Solar Power Europe. SRA. (2012). Solar Power Analysis and Design Speciﬁcations. Houston: Environmental Protection Agency. Stapleton, G. (2012). Grid Connected Solar Electric Systems. New York: Earthscan. Targosz, R. (2008). REDUCING ELECTRICITY NETWORK LOSSES. Retrieved Nov 1st, 2015, from http://www.leonardo-energy.org/reducing-electricity-network-losses Tipperary Energy Agency, 2014. Ireland’s Largest Photovoltaic Solar Panel Project is completed. Available at: http://tea.ie/irelands-largest-photovoltaic-solar-panelproject-is-completed/ Varatharajan, A. (2014). Harmonic Emission of Large PV Installations. Dresden, Germany: Renewable Energy and Power Quality Journal. WDC. (2010). Regional Enegy Balance & Biomass Heating Demand Estimates for 2020. Galway: Northen Periphery Program. WEC, W. E. (2007). Survey of Energy Resources. World Energy Council.

Hogan Sunderland Paper.indd 7

08/11/2016 14:33

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

Visit our website: buildingservicesnews.com Find us on Facebook

BSNews House Advert.indd 1

08/11/2016 14:36

MAIN SPONSOR

Best abstracts will be selected by a panel of assessors and a shortlist of entrants will be invited to submit full papers by September 2017. Final presentations will take place in October 2017 in DIT, Kevin Street. The competition is organised through the School of Electrical and Electronic Engineering.

The awards are open to all, with CIBSE, SLL and ILP members particularly encouraged to participate. Projects must be located in tIreland, while submissions can also be made which are based on lighting research. 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.

For the Irish Lighter Award, entries are encouraged from experienced lighting designers, or engineers who can present a paper about a finished project.

Irish Lighter

Servicesnews

Building

Michael McDonald, DIT. email: michael.mcdonald@dit.ie or Dr Kevin Kelly email: kevin.kelly@dit.ie

Who to contact

In the intervening years many others have followed in Ken’s footsteps and proudly represented Ireland at this annual event.

The Irish Young Lighter competition began in DIT in 2003 with the first students on the programme in Electrical Services Engineering. 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.

Young Irish Lighter

09:09

Lighter and Irish Young Lighter Awards – sponsored by CIBSE, organised by DIT and supported by SLL and Building Services News – are now being sought with preliminary 200/300-word abstracts required by June 2017.

The overall winners of the Irish Lighter and the Irish Young Lighter will each receive a cheque for €500, with prizes to each of the runner-up recipients in each competition.

20/10/2016

Following the very successful International Year of Light, submissions for both the Irish

Irish Lighter Awards 2017

Submissions invited for

Project1:Layout 1 Page 1

Institiúd Teicneolaíochta Átha Cliath Dublin Institute of Technology

School of Multidisciplinary Technologies College of Engineering & Built Environment The School of Multidisciplinary Technologies provides modules and programmes, at undergraduate and postgraduate levels, which link engineering and built environment disciplines for the design and operation of healthy, low-energy buildings and infrastructure for a modern sustainable world. These full-time and part-time courses are founded on a multidisciplinary research base and promote multidisciplinary themes across the DIT and with external partners. Themes include energy, sustainability, engineering computing (applied technology), building information modelling and management (BIM), and educational research.

Undergraduate Programmes Bachelor of Engineering (Hons) Engineering (General Entry)

Full Time

DT066

Bachelor Engineering Technology (Ord) Engineering (General Entry)

Full Time

DT097

MSc in Applied Building Information Modelling and Management (aBIMM)

Part Time

DT9876

Postgraduate Diploma in Collaborative BIM

Part Time

DT9876

Postgraduate Certificate in BIM Technologies

Part Time

DT9876

MSc in Applied Computing for Technologists

Part Time

DT093

FT/PT

Contact School

Check out our video “DIT Engineering General Entry” on YouTube.

Postgraduate Programmes

Check out our video “BIM@DIT Promotion” on YouTube.

Research Programmes PhD and MPhil opportunities

CPD By successfully completing modules in areas such as BIM and Applied Computing you can upskill in emerging areas and progress your career through the award of stand-alone CPD certificates or through the accumulation of credits into postgraduate awards. Review our available modules on: www.dit.ie/multidisciplinarytechnologies and www.dit.ie/bim

Please contact us if you have any queries about our programmes, research, or other activities. School Administrator: Jane Cullen. Tel: 01 – 402 4014. email: multidisciplinaryadm@dit.ie

Untitled-8 1

08/11/2016 14:28

A Reassessment of general lighting practice based on the MRSE concept

Christopher Cuttle MA Kit.cuttle@xtra.co.nz

Kit Kuttle Paper.indd 1

08/11/2016 14:31

SDAR Journal 2016

Abstract

1. The need for change

A case is made for reassessment of the purpose of general

The first professional lighting institution was founded in 1906 in New York under the slightly quaint title of The Illuminating Engineering Society, and this set the pattern for national and regional lighting institutions around the world. The general aim was to provide a sound, scientific basis for the development and application of electric lighting, and by any reasonable standards, those institutions have achieved notable success. From the outset, they faced the formidable task of making light a quantifiable commodity. To this day, light is the only one of the fundamental quantities defined by the General Council for Weights and Measures that is not specified purely in physical terms, but is actually defined in terms of human response. It was a major achievement when, in 1924, the International Commission on Illumination defined the lumen, relating human assessment of light to radiant power distribution, and this era has been described by the author as the first stage of the lighting profession1.

lighting practice, involving a change from lighting standards specifying illuminance for high levels of visual performance, to providing for predictable assessments of surrounding brightness. Mean room surface exitance (MRSE) is proposed as a suitable metric for this purpose. This metric actually serves a dual role, in that apart from providing practitioners with the means to design for chosen levels of surrounding brightness, it would enable regulators to specify for perceived adequacy of illumination, PAI. The adoption of PAI speciﬁed in terms of MRSE as the prime criterion for specifying indoor illumination levels in lighting standards would invoke fundamental changes in general lighting practice. These are discussed, together with limitations of the MRSE concept and the need for both further research and feedback from industry professionals.

The early approach to specifying provision of lighting was based on providing for peoples’ need for visibility, and the scientific community responded by introducing the concept of visual performance, which became the basis of general lighting practice. It was shown by research that speed and accuracy in detecting the detail of a visual task depends upon the angular size and luminance contrast of the critical detail, together with the illuminance incident on the task. In this way, by classifying the visual task difficulty associated with a broad range of human activities, lighting standards could be developed that specified minimum illuminance levels to perform specific visual tasks with speed and accuracy. At the same time, procedures for application were developed to enable compliance with the standards to be provided for with efficient use of resources. This scheme had every appearance of being a beautifully conceived application of scientific knowledge and engineering skill for the benefit of society at large, and while it may be seen as the second stage of the lighting profession1, its achievement has proved difficult. Since the end of the first stage, so much has been learned not only about human response to light, but about the role of human nature in lighting. It was in 1945 that the work of HC Weston2 at the National Physical Laboratories in the UK provided a research-based platform for developing lighting standards based on visual performance. However, this date happened to coincide with the onset of the proliferation of the fluorescent lamp, which caused lighting to no longer be thought of as a commodity to be applied stringently to provide for peoples’ needs, but as a means for generating feelings of wellbeing and stimulation. As if this was not enough, it soon became apparent that if something is found to be difficult to see, there are more effective ways of overcoming that than washing the visual task with light. Figure 1 shows an example of an office space lit by a combination of electric lighting and daylight, where the lighting distribution shows no pattern of relationship to visual tasks, but instead is directed towards providing for the appearance of a comfortable and pleasant environment. As in this instance, all around us we can see examples of the visual content of activities having been redesigned (screen-based reading tasks) or eliminated (bar-code readers), these innovations obviating the need for selective

Kit Kuttle Paper.indd 2

08/11/2016 14:31

A Reassessment of general lighting practice based on the MRSE concept

practice, in that MRSE would seem likely to provide a reliable indicator of surrounding brightness4, where this term relates to an overall assessment of how brightly-lit, or dimly-lit, a space appears to be. In doing so, he has explained, “This proposal is based on reason rather than research, and it is hoped that someone somewhere will feel motivated to investigate the validity of the concept for this purpose1”. Since then there has been both discussion and research concerning the MRSE concept, and it is time to evaluate the situation; but ﬁrst, a brief review of MRSE is in order.

Figure 1. An open-plan ofﬁce space, with separate enclosed meeting rooms, lit by a combination of electric lighting and daylight. The lighting distribution is unrelated to either visual tasks or the horizontal working plane, but is instead arranged to provide for a comfortable and pleasant working environment.

task lighting. It should be seen as remarkable that despite all these changes in how people interact visually with their surroundings that the level of illuminance on the horizontal working plane (HWP) persists as the metric that lighting practitioners employ for specifying illumination adequacy for all manner of human activities, irrespective of whether or not there is an identifiable visual task. This situation has not passed without challenge. It was once again New York that, during the 1960’s, took the lead with the formation of the Independent Association of Lighting Designers, IALD. This was to some extent in response to legal restrictions on the activities of the IES (it was registered as an educational institution), but also it was a reaction against the notion that the purpose of providing lighting was to be assessed in terms of satisfying prescribed illuminance values. This has led to a divided profession. On one hand, those who associate with illumination engineering institutions, such as IES and CIE, and on the other, those who associate with lighting design institutions, such as IALD and PLD. This has occurred despite several attempts to integrate engineering and artistic design objectives, of which perhaps the most notable was the ‘designed appearance method’ due to JM Waldram3, which sought to apply an illumination engineering approach for providing a designer-orientated distribution of lighting. While Waldram’s work gained significant accolades and would seem to have influenced some lighting designers, it failed to make any impact upon the course of general lighting practice. Instead, illuminance measured on the horizontal working plane persists as the universal metric for specifying illumination adequacy.

2. A proposal for change From the foregoing, the profession may be seen as continuing to specify lighting in terms of second stage objectives, while lighting practice has moved on to different design objectives. Third stage objectives, based on human response to light exposure, are yet to be addressed by general lighting practice. The author’s involvement with lighting practice, and his observation of the characteristics of lighting that are recognised as representing good current practice, have led him to propose mean room surface exitance (MRSE) as a better lighting metric for general lighting

2.1 The MRSE metric Mean room surface exitance may be applied in any enclosed space where inter-reflection between the surrounding surfaces generates a diffused light field, and crucially, it may be applied in two distinctly different ways. For lighting practitioners, MRSE may be used as a reliable means for indicating how peoples’ perceptions of surrounding brightness are likely to vary in response to lighting, regardless of the distribution of illumination, or of surface reflectances. For regulators, it would enable reliable specification of minimum illumination levels to satisfy the criterion of perceived adequacy of illumination, PAI, which indicates whether or not surrounding brightness at a specific location is perceived to be adequate for the human activity associated with that location. A procedure for predicting assessments of surrounding brightness would be directed towards characteristics of lighting distribution that are distinctly different from those employed in the familiar approach to assessment of lighting performance. It would be concerned with the density of luminous flux emanating from surrounding surfaces, rather than of flux incident upon them. This rules out illuminance as an appropriate metric, but it should not be assumed that attention is necessarily directed towards luminance. The visual effect to be characterised is an overall impression of the level of surrounding brightness, which does not depend (as luminance does) upon a particular viewing location or direction of view. Instead the form of measurement that is proposed is exitance, being the density of luminous flux (lm/m2) exiting, or emerging from, a surface. This line of reasoning leads to mean room surface exitance, MRSE, being the proposed metric for predicting the assessment of an adequately lit space. In this way, MRSE indicates both the average flux density emerging from surrounding room surfaces, and the average level of the diffused field of inter-reflected flux within the volume of a space. Understanding the manner in which the diffused light field is generated and sustained is crucial to recognising the workings of the MRSE metric. Within a room, MRSE is the average exitance of all room surfaces:

MRSE = Where

ΣAS MS = ΣASES ÚS ΣAS ΣAS

(1)

AS = area of surface S (m2) MS = exitance of surface S (lm/m2) ES = illuminance of surface S (lx) ÚS = reﬂectance of surface S

Kit Kuttle Paper.indd 3

08/11/2016 14:31

SDAR Journal 2016

Because an accurate calculation of MRSE involves determining the illuminance of every signiﬁcant surface within the room, this is best handled by a computer program. However, the author has also proposed an alternative procedure1 that has the attraction of not only being readily applied, but of making the workings of the procedure apparent, although it does incorporate an assumption that makes it less accurate: MRSE = Where

FRF ΣAS E(d)S ÚS = Aα ΣAS (1-ÚS )

(2)

FRF = First reﬂected ﬂux (lm) E(d)S = Direct illuminance of surface S (lx) Aα = Room absorption (m2)

Until the luminous flux emitted by the luminaires has undergone a reflection, it has no visible effect. The first reflected flux (FRF) is the source for the multiple inter-reflection process that generates the diffused light field within the volume of the space. For an enclosure of uniform surface reflectance, the average flux density within that field may be determined by application of Sumpner’s principle5, which states that as the total luminaire flux must equal the rate of flux absorption by the room surfaces, the average surface illuminance is given by dividing FRF by the room absorption, Aα, as indicated in formula (2). This provides a calculation procedure that can be carried out on the back of an envelope, and furthermore, the interrelationship between the characteristics of the room and the provision of light can be quite readily visualised. However, the assumption that, after the first reflection, all surfaces have reflectance values equal to the area-weighted average room surface reflectance value, inevitably introduces error. The extent of this error is discussed in Section 3, but as error is avoided by use of formula (1), it should be applied in all applications where accuracy is important. 2.2 Surrounding brightness It may be noted that some researchers (such as Rea et al, 2015) have made use of the terms ‘spatial brightness’ and ‘scene brightness’ to refer to brightness as presented to the eye. These terms have been defined in various ways by the different researchers, usually in terms of luminance distributions. The term ‘surrounding brightness’ is used throughout this paper to identify it as a distinct concept. Instead of being based on the notion of a scene presented to a viewer who is at a specific viewpoint and looking in a specified direction, it refers to an assessment of the overall brightness of an enclosed space, without regard to the viewer’s location or viewing direction. It is specified in terms of MRSE, and it is a response to an ambient condition, unaffected by body or head movement, and may be assessed on a multi-point scale of ‘very dim’ to ‘very bright’, as set out in the following subsection. Assessment of the adequacy of illumination may be seen as a step beyond brightness assessment. It is a judgement of whether or not the illumination is adequate for a specific purpose, and so it is a binary assessment for which the activity associated with the space is an influential factor. For example, a surrounding brightness level assessed as adequate for a doctor’s waiting room might be judged dim, or even gloomy, in the surgery. It is proposed that

the appropriate criterion for standards to regulate general lighting practice is the perceived adequacy of illumination, PAI, for which the corresponding level of surrounding brightness would depend upon the activity associated with the space. It should be noted that whereas brightness assessments can be obtained quite economically through use of laboratory viewing cabinets, data that enables comparisons of similar spaces but with different recognitions of associated activity calls for an altogether higher level of research commitment. Even so, this should be seen as a crucial research objective, without which, lighting standards are little more than iterations of commonly accepted practice. Based upon these considerations, MRSE is proposed as a metric to serve both types of assessment. It is proposed to fulfil the need for a metric that corresponds to typical human assessments of surrounding brightness, that is to say, how brightly-lit, or dimly-lit, a space appears to be. Also, it is proposed for specifying the perceived adequacy of illumination, PAI, for which assessment depends upon recognition that the space is associated with a specific human activity. While the extent to which MRSE fulfils these purposes has yet to be established, it is reasonable to assert that for any metric to do so, it would need to be some sort of measure that corresponds to the density of luminous flux from the surrounding surfaces that provide the stimulus for vision. On that basis, MRSE should prove to be a more appropriate metric than horizontal working plane illuminance. 2.3 MRSE research Various researchers have reported studies of human assessments of brightness, but too often the brightness levels are recorded in forms that are incompatible with the MRSE metric. Among the exceptions are a study by McKennan6, which is discussed in Section 3, and another by Rea, Mou and Bullough7. This latter study, which involved gathering responses from subjects exposed to controlled lighting conditions in a viewing cabinet, found better correlations between brightness assessments and illuminance at the eye on a vertical plane, than with horizontal illuminance. The author’s plea for “someone somewhere”1 to take up the challenge of investigating the MRSE concept led to research on this topic commencing at the Dublin Institute of Technology in 2011. James Duff has reported two experimental investigations involving human subjects, the first conducted in a laboratory viewing booth8 and the second in a small office9. In both situations, subjects assessed 27 lighting conditions, comprising three levels each of surrounding surface reflectances, luminaire flux distribution, and MRSE. Responses were recorded on the following scale: 7. Very bright 6. Bright 5. Slightly bright 4. Neither bright nor dim 3. Slightly dim 2. Dim 1. Very dim The key findings of the first study8 in a viewing booth were:

Kit Kuttle Paper.indd 4

08/11/2016 14:31

A Reassessment of general lighting practice based on the MRSE concept

•

A simple linear relationship was found to exist between MRSE and spatial brightness.

•

A broadly unpredictable relationship was found to exist between horizontal working plane illuminance and spatial brightness.

(As has been explained, the author prefers to use the term ‘surrounding brightness’ rather than ‘spatial brightness’ as the latter term has been defined in different ways by other researchers. However, the conditions of Duff’s experiments coincide well with the author’s definition of surrounding brightness which makes them directly comparable.) The second study9, in a full-scale office where the activity was readily recognisable, confirmed the above findings, and also included assessments of PAI. Again, a simple linear relationship to spatial brightness was found, and an additional finding was recorded: •

Levels of spatial brightness reported were strongly correlated with levels of PAI reported.

While conducting these investigations, Duff had to cope with various practical issues that were outside the range of conventional procedures. The measurement of MRSE involves gaining a response to the entire sphere of diffusely reflected light while ignoring direct flux from the luminaires. The difficulties he had to overcome led him to devise a novel procedure involving high dynamic range imaging, and he achieved this making use of available hardware and software10. He also examined calculation procedures for predicting MRSE, and investigated the extent of error incurred by formula (2), comparing MRSE values calculated by both formulae (1) and (2) for two different luminaire distributions, a downlighter and an uplighter, located at the centre of a room for which the five different reflectance combinations shown in Table 1 were specified11. Reflectance reflectance

Ceiling reﬂectance

Wall reﬂectance

Floor combination

0.5

0.6

0.5

0.4

0.7

0.5

0.3

0.8

0.5

0.2

0.9

0.5

0.1

Table 1. Reflectance combinations for Duff’s comparison11 of formulae (1) and (2), the results of which are shown in Figure 1. In every case the average room surface reflectance is 0.5, and the five combinations represent increasing levels of surface reflectance diversity.

The result of this comparison is shown in Figure 2. It can be seen that formula (2) tends to slightly underestimate MRSE due to downlighting, and, to a rather greater extent, to overestimate for uplighting. Luminaires that provide a balance of upward and downward ﬂux will incur errors between these levels, with the extent of error increasing as the diversity of reﬂectances increases. For practical applications, the underestimation incurred by using formula (2) will often be acceptable, as predictive calculations cannot be exact as they are liable to be upset, at least to the extent

Figure 2. Levels of error incurred using formula (2) rather than formula (1) in Duff’s comparison11 for downlight and uplight luminaires illuminating a room with the ﬁve reﬂectance combinations shown in Table 1.

indicated, by factors such as changes of furniture, to which MRSE would be more susceptible than horizontal illuminance. The higher level of error involved for uplighting is discussed in Section 3, but it may be noted that for luminaires that emit combinations of upward and downward flux, the actual error can be expected to fall between these extremes. While initial estimates of this sort can be instructive, for finalising installation specifications, Duff’s calculation procedure10 based on formula (1) should be applied.

3. Implications of proposed change The approach to lighting practice described in this paper involves changes in how lighting may be measured and calculated, and how it might be specified in standards. Underlying these practical changes is a fundamental difference of understanding as to what is the purpose of lighting. Instead of illuminating visual tasks to provide for visual performance, the prime purpose is understood to be to influence the appearance of overall brightness, or dimness, of the spaces that people occupy and use. Regulators would be able to specify lighting standards that would ensure that the people using a space would be likely to assess it to be adequately lit. Such standards would merely restrict lighting practitioners from providing lighting likely to be assessed inadequate, and so should not restrict how they choose to distribute light within the space, nor whether they opt to design for efficiency or for an illumination hierarchy. More generally, practitioners would be able to apply the MRSE concept to generate predicted assessments of surrounding brightness, and where standards do not apply, these could range from very dim to very bright, while they exercise full control over the distribution of lighting within the space. The shift from providing for visibility to providing for surrounding brightness is a fundamentally different understanding of the purpose of lighting, and its adoption for general lighting practice would cause practitioners to revaluate their current understanding of how their work influences human response. 3.1 MRSE and surrounding brightness The author has tentatively proposed1 a range of subjective assessments related to a logarithmic scale of MRSE, shown in Table 2. This was based on experience of practical measurements and student projects conducted over several years, but which fell short

Kit Kuttle Paper.indd 5

08/11/2016 14:31

SDAR Journal 2016

of the standards for publishable research. Duff’s experimental studies of the relationship between MRSE and human response represent the only research to date to meet that criterion, but the restricted scope of Duff’s experiments needs to be taken into account. It should be noted, for example, that Duff’s experimental situations exposed subjects to a range of three MRSE levels; 25, 50 and 100 lm/m2; and this covers only a small part of the scale indicated in Table 2. On the seven-point response scale shown in subsection 2.3, the subjects’ responses generally fell between 2 (dim) and 4 (neither dim nor bright), and although these assessments appear to accord reasonably well with the author’s descriptors, research studies covering a range of MRSE sufﬁcient to generate responses covering the entire range of responses, from very dim to very bright, are needed to provide acceptable conﬁrmation of the relationship. Mean room surface Perceived brightness or dimness 2 exitance (MRSE, lm/m ) of ambient illumination 10

Lowest level for reasonable colour discrimination

Dim appearance

100

Lowest level for ‘acceptably bright’ appearance

300

Bright appearance

1000

Distinctly bright appearance

Table 2. Tentatively proposed range of subjective assessments of lighting appearance related to mean room surface exitance1.

The data generated by Duff for his comparison11 of formulae for MRSE prediction provides insight into some practical differences from conventional practice that would be encountered in devising lighting installations to comply with MRSE standards based on surrounding brightness. Figure 3 shows a replotting of the data on which Figure 2 is based. In every case, the luminaire, whether a downlighter or uplighter, emits 5000 lumens, but the differences in MRSE levels produced could be expected to surprise experienced practitioners. It is the conventional understanding that while uplighting may produce attractive lighting effects, it is less efficient than downlighting and so should be reserved for applications where the purpose is to create decorative effects, and not used where efficiency is an important concern. This notion of ‘efficiency’ can be seen to be a direct consequence of the long-term effect of lighting practice being required to comply with horizontal working plane illuminance specifications, where the ‘efficient’ way to achieve compliance is inevitably to direct the luminaire flux onto that plane. Providing for surrounding brightness calls for a different way of thinking about what ‘efficiency’ means in lighting practice. Formula (2) shows the crucial role of first reflected flux FRF for generating MRSE. As the diversity of Duff’s five reflectance combinations increase, they follow the practice of conventional décor with higher levels of ceiling reflectance, and lower levels of floor reflectance values. The underlying principle is that when the purpose is to provide for surrounding brightness, ‘efficient’ application of luminous flux calls for the initial luminaire flux to be directed onto the room surfaces that have the highest reflectance. In this way, the pathway to efficient practice is not to unthinkingly direct light

Figure 3. Plot of MRSE levels based on Duff’s comparison11 of downlight and uplight luminaires calculated by exact formula (1) and approximate formula (2). The room characteristics are as described previously, and in every case the luminaire emits the same level of luminous ﬂux.

onto a specified measurement plane, but it is for the practitioner to start the process of devising an appropriate distribution of luminaire flux by evaluating the distribution of room surface reflectances. For conventionally decorated rooms, uplighting will be the optically efficient option, but if the ceiling is dark and the walls are light, then attention should switch to wallwashing. There is no such thing as a universally efficient luminaire. 3.2 Illumination hierarchy Even so, the pursuit of efficiency in conventionally decorated rooms would inevitably lead to successions of uplit rooms, all with softly diffused illumination reflected from matt white ceilings. While there are some spaces for which this type of lighting might be entirely appropriate, such as corridors, stairways and lift (or elevator) cars, there are far more spaces in which some surfaces or objects can be identified as deserving, or requiring, visual emphasis. Lighting practice that is directed towards compliance with current standards aims to achieve illuminance uniformity, but if standards were to be specified in terms of MRSE, then practitioners would have freedom to determine distributions of luminaire flux. Some might find this freedom confusing and opt for ‘design by rote’ solutions, but the very fact that practitioners would be able to comply with lighting standards while having the freedom to determine the distribution of direct flux, would open up opportunities in general lighting practice. It would enable practitioners, whether they consider themselves to be engineers or designers, to give consideration to the specifics of each space within a lighting proposal, and to develop an illumination hierarchy specific to each space, specified in terms of target/ ambient illuminance ratio, TAIR12,13. The author has proposed a scale relating TAIR to visual emphasis, shown in Table 3, and again, this is a proposal based on practical experience and student projects. It is yet to be subjected to rigorous research examination, but it should not be supposed that a relationship of this sort can ever be defined precisely. Its purpose would be to guide practitioners towards creating an ordered priority of visual emphasis related to the specifics of each individual installation. In this way, a practitioner would be able to make a statement by devising an illumination hierarchy that draws attention to selected objects and surfaces, whilst not being required to comply with a lighting standard. For this to become general practice, it would need the MRSE metric

Kit Kuttle Paper.indd 6

08/11/2016 14:31

A Reassessment of general lighting practice based on the MRSE concept

to become accepted by regulators, and TAIR to become accepted by practitioners. It may be expected that if such acceptance is achieved, these concepts would be taken up readily by lighting design software producers, who would see opportunities to extend the scope of their products into the lighting design process. Visual emphasis

Target/ambient Illuminance ratio, TAIR

Noticeable

1.5:1

Distinct

3:1

Strong

10:1

Emphatic

40:1

Table 3. Approximate guide to visual emphasis related to TAIR, being the ratio of target illuminance (the sum of direct illuminance and MRSE) to MRSE. (Adapted from Cuttle12).

For many practitioners, such acceptance would involve a reassessment of the purpose of lighting and procedures for its provision in general practice. The first level of understanding is that the flux from the luminaires travels through space without visible effect until it undergoes its first reflection, and this FRF becomes the source for both the MRSE (the diffused field of inter-reflected flux within the space), and the distribution of TAIR (which defines the illumination hierarchy). The next level of understanding concerns the two stages of optical control involved in achieving an illumination hierarchy. The luminaires that house the light sources provide the first stage of optical control by directing the distribution of initial flux, which is the source of FRF. The second stage of optical control is due to reflection from, and between, the objects and surfaces that comprise the lit space, and which become the second stage luminaire whose function is to present light to the users of the space. The author has published13 a spreadsheet that facilitates application of the concepts described in this paper. 3.3 The need for research The change of understanding that would follow from this reassessment of lighting would bring about changes in our perceptions of the limitations of our knowledge, and would generate a new set of priorities for researchers. Past studies of brightness have involved various terms to describe its appearance. It would be beneficial for researchers to adopt the seven-point brightness assessments scale used by Duff, and so enable comparisons between their findings. This scale avoids some confusions that have occurred in the past by involving just two descriptors – bright and dim. Some researchers have switched from dim to dark at the bottom end of the scale, but dark is the absence of light, and apart from astronomers seeking to retain their scotopic adaptation, the elimination of light has no place in lighting design. Emotive terms, such as gloomy and brilliant, should be avoided as they are context related, and as far as possible, researchers should avoid any form of implication that bright is good, or that dim is bad. For example, attractive displays of brightly lit objects in museums or retail premises (particularly for jewellery displays) may depend upon

the displayed objects being presented in settings that are dimly lit. Equally, evaluative terms such as acceptable, satisfactory or preferred, should be avoided. There would be plenty to occupy researchers in this new environment. In an earlier paper1 the author reviewed a study by McKennan6 in which he recorded overall brightness assessments as people moved between 16 differently lit spaces, and when they reached the end, they turned around and repeated their assessments going in the opposite direction. There was clear evidence that the assessment of each space was affected, significantly but not strongly, by the experience of the previous space. This suggests that MRSE specifications might need to take account of previous experience to give reliable indications of surrounding brightness. This discussion, and also the above study, have been restricted to enclosed spaces with electric lighting. There is no obvious reason why daylit spaces should not be treated similarly, but that needs to be verified. However, unenclosed spaces, as encountered outdoors, do not generate diffusely inter-reflected light fields, and so the MRSE concept would not be applicable. Even so, it should be expected that a metric that relates lighting to visual response would assess reflected light rather than incident light, and a move towards MRSE specifications for enclosed spaces should lead to research into suitable metrics for unenclosed spaces. Other discussion points raised by the author1,12,13 have included the effect of direct light at the eye, whether from luminaires or windows. It cannot be correct to add this stimulus when examining how light at the eye relates to assessment of surrounding brightness, but equally, it cannot be correct simply to ignore it, as MRSE does. It may be speculated that its effect would be to reduce surrounding brightness, particularly if it is strong enough to be a significant source of disability glare. Also, the concept of visual emphasis, which is a vital aspect of the illumination hierarchy concept, currently lacks any recognisable research basis. So while it is proposed that adoption of the concepts described in this paper would comprise a distinct step towards the third stage of the lighting profession, it should be expected that this step will open up new issues to be resolved, rather than solving the issues of lighting applications.

4. Conclusions While the MRSE approach indicates opportunities that have the potential to take general lighting practice a distinct step forward, it cannot be claimed that existing knowledge of the MRSE concept is sufﬁcient for it to be adopted for lighting standards to govern general lighting practice. Research to date does indicate that it is better suited for this purpose than horizontal illuminance, and so to that extent it should represent an improvement on current practice, but perhaps that is only because horizontal illuminance is so unsuited for the purpose. The fact is that more research is needed, but for that to occur, there needs to be an increased awareness of the potentials offered by a reassessment of the purpose of general lighting practice.

Kit Kuttle Paper.indd 7

08/11/2016 14:31

SDAR Journal 2016

Acknowledgements

References

The author thanks Dr Kevin Kelly for his advice and comments on this paper, and James Duff for the use of his material. He also thanks the three anonymous reviewers for their comments, which provided helpful contributions to the ﬁnal review of the paper. Figure 1 was photographed by Paul McCredie, in the architectural practice of Stephenson and Turner, Wellington, New Zealand.

1. Cuttle C. Towards the third stage of the lighting profession. Lighting Research & Technology 2010; 42(1): 73-93. 2. Weston HC. The relation between illumination and visual efficiency – The effect of brightness contrast. London: His Majesty’s Stationery Office, 1945. 3. Waldram JM. Studies in interior lighting. Transactions of the Illuminating Engineering Society (London) 1954; 19: 95-133. 4. Cuttle C. Lighting by Design, London: Routledge, 2003. Second edition, 2008. 5. Cuttle C. Sumpner’s principle: A discussion. Lighting Research & Technology 1991; 23(2): 99-106. 6. McKennan GT. A study of the sequential experience of different lighting. Lighting Research & Technology 1981; 13(1): 1-10. 7. Rea MS, Mou X, Bullough JD. Scene brightness of illuminated interiors. Lighting Research & Technology 2015, doi: 10.1177/1477153515581412 8. Duff J, Kelly K, Cuttle C. Spatial brightness, horizontal illuminance and mean room surface exitance in a lighting booth. Lighting Research & Technology 2015, doi: 10.1177/1477153515597733. 9. Duff J, Kelly K, Cuttle C. Perceived adequacy of illumination, spatial brightness, horizontal illuminance and mean room surface exitance in a small office. Lighting Research & Technology 2015, doi: 10.1177/ 1477153515599189. 10. Duff J, Antonutto G, Torres S. On calculation and measurement of mean room surface exitance. Lighting Research & Technology 2016; 48(3): 384-388. 11. Duff JT. Research note: On the magnitude of error in the calculation of mean room surface exitance. Lighting Research & Technology 2016, doi: 10.1177/1477153516659519. 12. Cuttle C. A new direction for general lighting practice. Lighting Research & Technology2013; 45; 22-39. 13. Cuttle C. Lighting Design: A perception-based approach. Oxford: Routledge. 2015.

Kit Kuttle Paper.indd 8

08/11/2016 14:31

IBC DIT Bulid Serv 2016.indd 1

08/11/2016 14:23

DIT offers PhD opportunities in Engineering and the Built Environment - wholetime and part-time; funded, self-funded or partnership funded with companies. To find out about current opportunities contact marek.rebow@dit.ie

OBC PhD Advert.indd 1

www.dit.ie

12/11/2016 09:58