Technology Customer Service 600, avenue de la Montagne, Shawinigan (Québec) Canada G9N 7N5

Transcription

1 Technical Report LED street lighting in the municipality of Saint-Gédéon-de- Beauce within the framework of advanced lighting technologies LTE-RT Public distribution May 2011 André Laperrière Technology Customer Service 600, avenue de la Montagne, Shawinigan (Québec) Canada G9N 7N5

2 LED street lighting in the municipality of Saint-Gédéon-de-Beauce within the framework of advanced lighting technologies Copy no Author: André Laperrière Collaborators: Chrisnel Blot, Spectralux Éric Lachance, Saint-Gédéon-de-Beauce Pierrette Leblanc, Natural Resources Canada Patrick Martineau, Hydro-Québec Project manager: André Laperrière Completed within the framework of the project: Advanced lighting technologies J-4024 Applicant: Plateforme Clientèle Project manager of the business unit: Patrick Martineau Approved by: Jocelyn Millette Manager Technology Customer Service Hydro-Québec Research Institute

3 LIST OF INDIVIDUALS OR GROUPS WITH ACCESS TO THE DOCUMENT COMPLETE REPORT: COPY N Jocelyn Millette Manager Technology Customer Service 1 My Dung Handfield - Chef Technology Customer Service 2 Michel Dostie Leader Expertise Energy Use 3 Éric Dumont Manager, Building Energy 4 André Laperrière - Researcher LTE 5 Éric Lachance Mayor, Saint-Gédéon-de-Beauce 6+PDF Guy Courcelle LDI Technology group 7 Pierrette Leblanc Natural Resources Canada 8+PDF Roger Bellemare S.C.U.E. Hydro-Québec 9 Patrick Martineau S.C.U.E. - Hydro-Québec 10 Omer Lemay S.C.U.E. Hydro-Québec 11 Chrisnel Blot Spectralux 12 Nathalie Blanchard, Optical Design, INO 13 Lighting Committee Hydro-Québec (site Livelink) PDF LTE library (Original copy) advanced lighting technologies iii

4

5 Summary Within the framework of lighting technologies, LED technology is beginning to offer energy savings opportunities for certain markets. However, like any other new technology, it is important to fully understand the way it operates, its physical principles, its limits, etc. Regarding road lighting, several governing bodies, such as the DOE (Department of Energy) in the US have started to promote the technology. However, some studies have tended to show that LED technology must be used with caution and that energy savings are not as great as has been suggested in all the cases. A pilot project was implemented in the municipality of Saint-Gédéon-de-Beauce, and laboratory tests were concurrently executed by the company Spectralux. During these tests, photometric, colorimetric and electric aspects were studied, including mesopic correction and night vision, simultaneously for both LED technology and conventional High Pressure Sodium (HPS) lighting. This is a new concept according to which the vision is different at night (scotopic or S) at low-level light than during the day (photopic or P) and consequently, it is possible to adapt the spectral separation in order to optimize energy savings. In the traditional method, all the calculations are done on the basis of daylight correction, i.e. photopic correction, when in reality we should be using mesopic correction, i.e. a correction that lies between the photopic and scotopic ones. Table S-1: Comparison of LED technology and HPS by type of vision Photopic (daylight vision) (lumens) Scotopic (night vision) (lumens) S/P ratio LED luminaire HPS luminaire (ballast factor of 1) In addition, a survey was conducted by the municipality of Saint-Gédéon-de-Beauce regarding the pilot project. The analysis shows that it is possible to reduce the consumption from 130 watts (100 watts HPS lamps) to 65 watts per luminaire with the LED technology. However, there is a reduction in the level of lighting when compared to the HPS technology. In the case of local streets, the levels of brightness were nevertheless shown to be sufficient. Overall, the laboratory tests that included digital simulations confirm the level of performance obtained on site. advanced lighting technologies v

6 Figure S-1 shows a comparison of the two technologies (LED 4100 K and HPS) for a local road, i.e. a residential road 24 feet wide (2 lanes) where lamp posts are 150 feet apart and 30 feet high. According to the IES RP-8 standard, a luminance level of 0.3 cd/m 2 is recommended for local roads with light road traffic. The simulations show that the technology would be able to meet adequate performance for certain types of application. It should be noted that technology is evolving rapidly and that new markets will increasingly become available. We should exercise caution and therefore be meticulous when drawing conclusions. Today, new products already offer improved performance when compared to the products installed at the beginning of the pilot project. Figures S-2 and S-3 show the distribution of lumens by luminaire, for both LED 4100 K and HPS. It can be observed that in the case of HPS luminaire, we obtain lumens on the street side with 130 watts, while we obtain lumens with LED technology for 65 watts. We remind that one lux is equal to one lumen per square meter. It is interesting to note that a LED luminaire produces the same photometric performance even if the electric current varies by + / - 10%, while this is not the case with the HPS luminaire with a magnetic ballast. Lastly, it should be noted that a European standard of the International Commission on Illumination (CIE) recommends levels of lighting even lower than those prescribed by the North American RP-8 standard. This aspect is analyzed in the present report. advanced lighting technologies vi

7 HPS LED 0.65 cd/m 2 left side 0.56 cd/m 2 right side 0.45 cd/m 2 left side 0.31 cd/m 2 right side Mesopic correction y = x x x x represents photopic luminance MULTIPLIER= 1.13 Mesopic correction y = x x x MULTIPLIER = cd/m 2 left side 0.39 cd/m 2 right side Figure S-1: Calculated initial levels of luminance on local street with 2 lanes (24 feet), with street lamps 150 feet apart and 30 feet high (4100 K) advanced lighting technologies vii

8 New lamp reference ballast: lumens Luminaire output (BF = 0.9): lumens (Luminaire efficiency: 64.2%) Lumens downward: lumens Lumens upward: 299 lumens Lumens on the house side toward the ground lumens lumens Lumens street side toward the ground Figure S-2: Distribution of lumens, HPS luminaire 100 watts lamp (total of 130 watts) advanced lighting technologies viii

9 Total of lumens: lumens Lumens on the house side toward the ground lumens lumens Lumens street side toward the ground Figure S-3: Distribution of lumens, LED luminaire (total of 65 watts) Using figure S-1, we can determine the efficiency for a street with two lanes: Table S-2: Summary of luminance by side of the lane and technology Power Right side Left side (watts) cd/m 2 cd/m 2 LED HPS advanced lighting technologies ix

10 Efficiency in terms of cd/m 2 per Watt can therefore be calculated by dividing average luminance by electricity consumption. Table S-3: Efficiency by technology in cd/m 2 per Watt of electricity Power Efficiency right side Efficiency left side (watts) cd/m 2 per Watt cd/m 2 per Watt LED HPS Following this approach, we can therefore obtain the efficiency ratio, changing from HPS to LED. Table S-4: Increase in efficiency when changing from HPS to LED Right side ratio Left side ratio LED/HPS RATIO If the average of the two lanes is calculated, the efficiency increases by 1.5, or close to 50%. Annual energy savings, in monetary terms, come to $23.69 per street luminaire, for a reduction in power of 65 watts, from 130 watts to 65 watts. Lastly, it is hoped that the present report will allow the readers to make an informed choice concerning new advanced lighting technologies for road lighting. André Laperrière, Researcher Technology Customer Service Energy Technology Laboratory (L.T.E.) advanced lighting technologies x

11 Acknowledgments Principal author, André Laperrière, would like to thank all the individuals who contributed to the preparation of this report, including the municipality of Saint-Gédéon-de-Beauce. Lastly, I would like to acknowledge the Natural Resources Canada s financial contribution to this project. advanced lighting technologies xi

12

13 Table of contents Pages INTRODUCTION ROAD LIGHTING AND ITS PRINCIPLES Illuminance and luminance Illuminance: Luminance: Recommended luminance values Lighting calculation grid Pavement types Street classification CIE 115:2010 STANDARD: LIGHTING OF ROADS FOR MOTOR AND PEDESTRIAN TRAFFIC PROJECT OVERVIEW FEATURES OF LIGHT-EMITTING DIODES Life span Colour rendering index Correlated colour temperature (CCT) SEQUENCE OF TESTS FOR THE NEW LIGHT-EMITTING DIODE TECHNOLOGY Measurements of diode luminaires in integrating sphere Absolute photometry Astro with 6000 K diodes Absolute photometry Astro with 4100 K diodes TESTS ON THE 5 TH AND 8 TH STREET th Street th Street Illuminance calculations for conventional HPS cobra-style luminaire Illuminance calculations with 4100 K LED luminaire Illuminance calculations with a 4100 K LED luminaire on 5 th Street Illuminance calculations with HPS luminaire on 5 th Street MESOPIC CORRECTION FOR LED LUMINAIRE MESOPIC CORRECTION FOR HPS LUMINAIRE ASSIST AND LED VERSUS HPS ASTRO 6000 K VERSUS ASTRO 4100 K STREET LUMINAIRE EFFICIENCY CONCLUSION...53 ANNEX A: SPHERE TESTS FOR 4100 K LED LUMINAIRE (L C1)...55 ANNEX B: PHOTOMETER TESTS FOR 4100 K LED LUMINAIRE (S R1)...60 ANNEX C: PHOTOGRAPHS OF THE MUNICIPALITY OF SAINT-GÉDÉON-DE-BEAUCE...72 ANNEX D: REQUIRED ROADWAY LIGHTING LEVELS BY THE CITY OF OTTAWA FOR URBAN AREA...79 ANNEX E: SURVEY DISTRIBUTED TO THE MUNICIPALITY...81 advanced lighting technologies xiii

14 ANNEX F: SITE MEASUREMENTS OF THE OPERATING STATE...83 ANNEX G: SITE MEASUREMENTS OF LED LUMINAIRE LIGHTING...85 ANNEX H: ENERGY SAVINGS...87 ANNEX I: SPECIFICATIONS OF THE CITY OF LOS ANGELES...89 advanced lighting technologies xiv

15 List of figures Pages Figure 1: Principle of light measurement (lux)... 5 Figure 2: Angles of the observer for illuminance and luminance calculation... 6 Figure 3: Angles of the observer for the calculation of veiling luminance L v... 7 Figure 4: Calculation grid for the space between two (2) luminaires... 9 Figure 5: Calculation grid according to IES RP Figure 15: Correlated colour temperature Figure 16: Overall efficiency for an HPS luminaire in clean condition and new lamp reference ballast (ballast factor of 1) Figure 17: Overall efficiency of an HPS luminaire in clean condition and new lamp ballast factor of Figure 18: Lumens distribution for a 4100 K LED luminaire (65 watts total) Figure 19: Spectral power distribution of LED illuminance for the pilot project Figure 20: Distribution of luminous flux by type of vision LED 4100 K Figure 21: HPS luminaire installed inside the sphere Figure 22: Spectral power distribution for HPS luminaire Figure 23: Spectral distribution for HPS luminaire test L C Figure 24: Distribution of luminous flux by vision type - HPS Figure 25: Mesopic luminance / photopic luminance Figure 26: LED mesopic luminance / HPS mesopic luminance Figure 27: Comparison of colour temperatures of LED luminaires Figure 28: Spill light Figure 29: Downward efficiency and total efficiency advanced lighting technologies xv

16

17 List of tables Pages Table 1: Levels of lighting according to IES RP Table 2: Recommended values of luminance ratios and luminance Table 3: CIE 115:2010 Lighting of roads for motor and pedestrian traffic and P Class Table 14: Measurements of LED luminaires in integrating sphere Table 15: Astro 6000 K diodes on goniophotometer Table 16: Astro with 4100 K diodes on goniophotometer Table 17: Used HPS luminaire in clean condition and a new lamp on reference ballast Table 18: New HPS lamp on reference ballast Table 19: Simulation results for 5 th Street with 150 feet of spacing between luminaires and 4100 K LED luminaire Table 20: Simulation results for 5 th Street with 170 feet and 190 feet of spacing between 4100 K LED luminaires Table 21: Simulation results for 5 th Street with 150 feet of spacing between luminaires and HPS luminaire Table 22: Lumens measured in sphere, LDI LED 4100 K luminaires Table 23: Photopic and scotopic lumens by wavelength, LDI LED 4100 K luminaire Table 24: Summary of measurements in integrating sphere, LED 4100 K luminaire Table 25: Summary of measurements in integrating sphere for HPS luminaires Table 26: Lumens measured in integrating sphere, HPS luminaire / reference ballast Table 27: Photopic and scotopic lumens according to wavelength, HPS luminaire Table 28: LED / HPS ratio by luminance level Table 29: Summary of luminance by side of the lane and technology (including LED mesopic correction) Table 30: Efficiency by technology in cd/m 2 per electric watt Table 31: Gain in efficiency resulting from the replacement of HPS by LED advanced lighting technologies xvii

18 Introduction Within the context of road lighting, the new LED technology is slowly making its way and is increasingly garnering interest. Consumers are increasingly turning to technology in order to reduce their energy costs but also to meet the necessary performance standards. It is in this context that a pilot project took place in the municipality of Saint-Gédéon-de-Beauce, by using LED luminaires of the company LDI. With the intention of evaluating this new technology on site, an experiment procedure was developed. The luminaires were first evaluated in the laboratory by the company Spectralux from Montreal. During this stage, the tests took place in the laboratory on the integrating sphere as well as on a goniophotometer. This approach was executed by using both the current High Pressure Sodium (HPS) technology and the LED technology. Simulation tests were then conducted with the help of the software Visual Roadway Lighting Tool. Following this approach and this technological exploration over a period of time, tests were conducted on site in order to validate the levels of lighting obtained through simulation. This report has for the objective to gain an understanding of the new technology in a very specific case which is residential road lighting. The practical objective of the investigation is to determine the potential energy savings, all the while ensuring the lighting quality obtained. advanced lighting technologies 1

19

20 1. Road lighting and its principles The IES-RP-8 standard is used for road lighting, and it is deemed necessary to understand the principles of calculation. The street type termed local corresponds to the residential sector. Table 1: Levels of lighting according to IES RP-8 Road and Pedestrian Conflict Area Route Pedestrian Conflict Area R2 & R3 lux Uniformity ratio E avg / E min (Max allowed) Veiling luminance ratio L Vmax / L avg (Max allowed) Freeway Class A Freeway Class B High Expressway Medium Low High Major Medium Low High Collector Medium Low High Local Medium Low advanced lighting technologies 3

21 The IESNA RP-8 standard describes the lighting levels in lux according to the type of roadway surface, type of road as well as the pedestrian conflict area, which is defined as the pedestrian activities in relation to the number of pedestrians per hour: High: Medium: Low: 100 or more pedestrians per hour 11 to 99 pedestrians per hour 10 or fewer pedestrians per hour In general, the classification R3 is used for asphalt roadways, which is due to the difference in material reflectance. According to the IES RP-8 criteria, the method of recommended values according to the luminance method will be presented later. In order to fully understand the difference between these two methods, lighting ( illuminance ) and luminance, it is important to refer to the definition: The density of the luminous flux on one point of a surface is defined as the luminous flux per unit area. 1.1 Illuminance and luminance Illuminance: The density of the luminous flux (E h ) is also known as illuminance or lighting level. The SI unit of illuminance is lux (lx) where 1 lux = 1 lumens / m 2. As illustrated in the figure below, a photometer is used to measure the lighting level. It should be noted that the lighting level measured is independent of the reflection of the surface. E h d da advanced lighting technologies 4

22 1 Figure 1: Principle of light measurement (lux) Given that E h represents horizontal light, I being the luminous intensity in cd/m 2, it becomes possible to calculate light using ISL (Inverse Square Law) as a function of angles and shown in figure 2: E h I(, ) Cos( ) LLF 2 D where and becomes: E h D H Cos( ) I(, ) Cos H 3 2 ( ) LLF LLF represents the Light Loss Factor, i.e. a factor associated with luminous depreciation as a function of time through source burnout, dirt depreciation, etc. 1 advanced lighting technologies 5

23 Figure 2: Angles of the observer for illuminance and luminance calculation Luminance: Luminance represents the amount of light that is reflected by a surface and that is perceived by the eye of an observer. It is therefore the amount of light perceived by the eye of an observer. While, traditionally, the IES standards used the illuminance method, the preferred method now is the luminance method that takes further into account the surface environment and type. The observer is located at a distance of metres from the fixation point, the eyes are at the height of 1.45 metres from the street, and the observer is looking from an angle of 1 from the horizontal of the surface. 2 2 ROADWAY LIGHTING DESIGN METHODOLOGY AND EVALUATION; Olkan Cuvalci (Western Kentucky University Engineering Technology Department Kentucky); Bugra Ertas (A&M University Mechanical Engineering Department Turbomachinery Laboratory College Station, Texas), 2000 Society for Design and Process Science advanced lighting technologies 6

24 Figure 3: Angles of the observer for the calculation of veiling luminance L v Luminance is calculated in the following manner, r being the reflection of the road: L p n i 1 r( ) I(, ) i, i H i 2 i L p refers to figure 2 Lastly, D 2 ( H e) ( b o ) 2 D refers to figure 3 advanced lighting technologies 7

25 The luminaire emits light directly to the eye of the observer and consequently produces a reduction in the visual performance and a feeling of discomfort. Due to this sensation, luminance can be higher than the actual luminance coming from the reflection of light on the road surface. This veiling luminance can be calculated empirically in the following way: L v 10 Ev 2 1, 5 Ev being the vertical illuminance on the plane of the observer s pupil the angle between the line of sight and luminaire in degrees The IES method defines the veiling luminance ratio as the maximum value of veiling luminance divided by the average pavement luminance as a measure of the disability glare or the discomfort produced by the glare. 1.2 Recommended luminance values Between two luminaires A and B, it is possible to obtain the following values for each one for the following twenty (20) points, for example. The concept that needs to be understood is that it is possible to determine an average value as well as longitudinal uniformity on a grid. We can imagine a situation where the average value would be higher but the distribution would be very poor. Consequently, some points would be dark holes or some would be too bright, pointing to an inadequate optimization of the lighting system. advanced lighting technologies 8

26 Figure 4: Calculation grid for the space between two (2) luminaires L avg luminance average between the two (2) luminaires Lmin Lmax luminance minimum between the (2) luminaires luminance maximum between the two (2) luminaires Lv veiling luminance; Lv max maximum veiling luminance It should be noted that it can be difficult to measure luminance on site, while this is not the case for illuminance. Consequently, measurements on site are often done using lux meter and illuminance. Table 2 shows required luminance average as well as uniformity ratios according to different types of road and road traffic rates. One of the points of interest in this project is the road lighting of local, i.e. residential, type with low pedestrian conflict for which required average luminance is 0.3 cd/m 2. advanced lighting technologies 9

27 Table 2: Recommended values of luminance ratios and luminance 3 Road and Pedestrian Conflict Area Route Pedestrian Conflict Area Average luminance L avg (cd/m 2 ) Uniformity ratio L avg / L min (Maximum allowed) Uniformity ratio L max / L min (Maximum allowed) Veiling luminance ratio L Vmax / L avg (Maximum allowed) Freeway Class A Freeway Class B High Expressway Medium Low High Major Medium Low High Collector Medium Low High Local Medium Low In a sector of low pedestrian conflict, the following luminance characteristics are necessary. In North American, illuminance (lux) values were traditionally used; however, luminance criteria (cd/m 2 ) are now being used. 3 Reference IES RP-8 advanced lighting technologies 10

28 1) CRITERIUM 1: L avg greater than 0.3 cd/m 2 This ensures that the level of luminance of the roadway is sufficient. When the spacing is too great, the level of luminance is not adequate. However, the level of luminance increases with the size of the road and the higher pedestrian conflict area. L avg < 6.0 2) CRITERIUM 2: L min This criterion targets the uniformity of luminance. If the minimum luminance is too low, the ratio becomes infinity. In such a case, we get: L L min avg L avg 0 Lmax 3) CRITERIUM 3: Lmin < 10.0 This criterion ensures luminance uniformity and a maximum ratio between the maximum value and the minimum value of luminance. L v 4) CRITERIUM 4: max < 0.4 Lavg Lv Veiling luminance; Lv max being the maximum veiling luminance. If the maximum veiling luminance is greater than 40% of the average luminance value, then the produced glare is disturbing and blinding. The veiling luminance is added to the luminance produced by the light reflecting on the surface. Therefore, there are four (4) criteria that need to be taken into consideration for road lighting. Overall, not only does the luminance level need to be considered, but also i) uniformity levels and ii) the glare levels. 1.3 Lighting calculation grid The figure below illustrates the points of the calculation grid recommended by IESNA and the location of the observer in the luminance measurements. advanced lighting technologies 11

29 Figure 5: Calculation grid according to IES RP-8 This calculation grid is issued by IESNA in order to ensure that the calculated levels of lighting correspond to the same points during comparative studies of photometric performance of luminaires. The statistical values (average, maximum, minimum) vary according to the location of the grid points. These same grid points are used in the calculations of illuminance, luminance, and small target visibility (STV) on the roadway. For regular and straight sections, a cycle with ten points between the two posts is used with two grid points per traffic lane. A luminaire cycle is the spacing between two successive luminaires on the same side of the road. For the illuminance calculations, it is recommended to use a minimum of three luminaire cycles, that is, in the test area, before the test area and after the test area. In contrast to what is recommended in the text under figure 5, for luminance measurements and the small target visibility, Lighting Analysts recommends to use a minimum of five luminaire cycles, that is, in the test area plus a cycle before the test area and three cycles after the test area. The luminance levels vary according to the number of luminaires. advanced lighting technologies 12

30 The width of a traffic lane is 3.65 meters (12 feet). Transversely (perpendicular to the roadway), the first grid point of each traffic lane is ¼ of the width of the traffic lane and the increment between the grid points is ½ the width of the lane, or meters (6 feet). Longitudinally (direction of the traffic), the first grid point of each traffic lane is 1/20 of the spacing between the luminaires and the increment between two consecutive points of the calculation grid is equal to 1/10 of the spacing between the luminaires and must not exceed 5 meters. 1.4 Pavement types IESNA defines four types of pavement according to the materials used and the mode of reflectance. The data on the different types of pavement are provided in Table 1 on page 5 of the IESNA recommendation RP R1 pavement is cement with a reflectance coefficient greater than 10%. R2 pavement is asphalt composed of 60% gravel greater than 1 cm with a reflectance coefficient of 7%. R3 pavement is asphalt composed of dark aggregates with a reflectance coefficient of 7%. In North America, this type of pavement is used the most. R4 pavement is asphalt with very smooth texture and a reflectance coefficient of 8%. advanced lighting technologies 13

31 1.5 Street classification According to the Illuminating Engineering Society of North America (IESNA), streets are classified as major, collector and local. It should be noted that the residential terminology used in this context is not linked to the number of residences or houses located on the street. From the geographic point of view, an arterial road crosses through the entire city; a collector street flows into an arterial road, while a local street flows into a collector street. In terms of lighting, annual average daily traffic (AADF) per hour and the pedestrian activity are two parameters used by IESNA to classify a road. Annual average daily traffic Annual average daily traffic (AADF) is an estimate of the number of vehicles that circulate on a street in any given year divided by the number of days in the year. This flow is calculated for both traffic directions. The AADF is calculated using a statistical method for estimation applied to about collector streets at more than sites, bringing the total to about automatically recorded days. The observed periods and the frequency vary greatly, from several days at temporary sites to an entire year at permanent sites. If the AADF per day is greater than vehicles, the road is classified as an arterial road. If the annual average daily traffic flow per hour is between and vehicles, the street is classified as collector. The street is classified as local if the AADF per hour is between 100 and vehicles. Pedestrian activity Pedestrian activity is defined as the level of car/pedestrian interference. Pedestrian activity is characterized as low when only a few pedestrians (10 or fewer) tend to cross the street occasionally at nighttime. Pedestrian activity is said to be medium when a larger number of pedestrians (between 11 and 99) cross the street occasionally at nighttime. It is said to be high when many pedestrians (100 and more) cross the street frequently at nighttime. The number of pedestrians should be counted at nighttime for a given period of one hour on a typical block or over a distance of 200 meters on both sides of the street at the busiest time (generally between 6:00pm and 7:00pm, but the actual time could however vary from city to city or from neighbourhood to neighbourhood). advanced lighting technologies 14

32 Periods that are particularly busy, such as the moment when crowds are exiting a movie theatre, entertainment venues or sport events, should be noted, as well as when bars and stores are closing at night. Areas that are reserved for pedestrians at intersections or other places constitute a special case that is referred to as pedestrian conflict as opposed to pedestrian activity. Therefore, the level of vehicle/pedestrian interference is associated with each of the three street classifications (arterial, collector and local) in order to determine the required level of lighting. For example, for the arterial classification, there are three levels of car/pedestrian interference, as well as for collector and local. advanced lighting technologies 15

33

34 2. CIE 115:2010 Standard: Lighting of Roads for Motor and Pedestrian Traffic The technical report CIE 115:2010 Lighting of Roads for Motor and Pedestrian Traffic by the International Commission on Illumination is a 2010 revision of the 1995 report. It takes into account additional factors such as energy efficiency and control systems with the goal of reducing lighting levels during periods of reduced activity. The systems are divided into three categories: M, C and P. Different concepts are used for calculation purposes: 4 Motorised traffic, M, (for drivers of motorised vehicles luminance) Conflict areas, C, (where traffic streams intersect, or run into areas with pedestrians, cyclists, or there is change in geometry or parking areas luminance or illuminance) Pedestrian and low speed areas, P, ( for needs of pedestrians illuminance, H and V) In the context of the pilot project, the concept of interest is type P, residential sector or pedestrian and low speed area. Table 3: CIE 115:2010 Lighting of roads for motor and pedestrian traffic and P Class Cat. Average horizontal illumination E h, av (lux) Minimum horizontal illumination E h,min (lux) Additional criteria if facial recognition is necessary Minimum vertical illumination E (lux) v,min Minimum vertical semi-cylindrical illumination E (lux) P P P P P P sc,min 4 CIE and Roadlighting, Steve Jenkins, Division 4 Representative advanced lighting technologies 17

35

36 3. Project overview The objective of the present report is to conduct in close collaboration with Spectralux and Hydro- Québec a theoretical and experimental analysis of the conventional, high pressure pressure, cobrastyle luminaire technology and the new, light-emitting diode luminaire technology. The experiments, measurements, analyses and were conducted jointly in partnership with Hydro- Québec and Spectralux laboratory. The tests were carried out on the following: absolute photometry of 3 cobra-style luminaires, in a dirty condition, 100 watts, high pressure pressure, removed from the site; absolute photometry of 3 cobra-style luminaires, in a clean condition, 100 watts, high pressure pressure, removed from the site; characterization of three used, high pressure pressure lamps of 100 watts used in the cobra-style luminaires with reference ballast; characterization of three used, high pressure pressure lamps of 100 watts used in luminaires with commercial ballast from a cobra-style luminaire; characterization of three new, high pressure pressure lamps of 100 watts in integrating sphere with reference ballast; absolute photometry of 3 Astro luminaires of 60 watts with light-emitting diodes of 6000 K; absolute photometry of 3 Astro luminaires of 60 watts with light-emitting diodes of 4100 K; lighting simulations using photometries of 100 Watt, high pressure pressure luminaires and 60 Watt luminaires with light-emitting diodes of 6000 K et 4100 K; establishing a comparison of performances of the two models of luminaires. The report elaborates on a variety of road lighting notions which professionals working in the lighting field may find redundant. However, given that the lighting field is a new area for diode manufacturers, it is important to clarify all the concepts and provide a complete presentation of the diode. In this sense, the present report also has an educational role and contributes to the transfer of knowledge. advanced lighting technologies 19

37 4. Features of light-emitting diodes The luminous flux of a diode and its reverse voltage Vf are characterized by a nominal current published by the manufacturers. With these two features, it is possible to determine the pair (luminous flux, power) in order to establish luminous efficacy in lumens per watt. 4.1 Life span A light-emitting diode does not operate in the same way as a traditional luminous source: it can last for an extremely long period of time. This explains why, at the beginning of the 2000s, diode manufacturers were stating that the overall life span of diodes was hours. However, the quantity of the luminous flux decreases with time, so that after a certain threshold a diode is considered to be inoperative. Today, diode manufacturers publish shorter life spans, around hours. Diode life span is generally defined as the time it takes for half of a group of diodes to emit less than 70% of the initial flux. This performance criterion is called L70. This information is generally provided by diode manufacturers. 4.2 Colour rendering index The colour rendering index measures the ability of a light source to reproduce colours of objects it illuminates. The reference source, with a colour rendering index of 100, is an incandescent lamp which is considered to be a black body. Initially, the International Commission on Illumination (CIE) defined the colour rendering index as a capacity of a light source to reproduce 14 colours. Since fluorescent lamps do not reproduce well red hues, the CIE revised its method to reduce the number of colours from 14 to 8. Due to the fact that calculated colour rendering indexes do not adequately describe the visual perception of colour rendering by white light-emitting diodes, the CIE is currently looking for ways to implement a new metrics which would be adapted to diodes. advanced lighting technologies 20

38 4.3 Correlated colour temperature (CCT) The temperature of a light source characterizes the hue of the white colour. The light is defined as cold if the CCT value is high and warm if the CCT is low. Higher CCT values represent bluish tinge, while lower CCT values represent yellowish tinge. Temperature colours and values CIE chromaticity diagram Figure 15: Correlated colour temperature On the diagram CIE 1931 above, pure colours (monochromatic radiations) are located along the boundary curve, while mixed colours are inside the diagram and the white colour is in the centre. The straight line which links the two extreme points of the spectrum is called the purple line. advanced lighting technologies 21

39

40 5. Sequence of tests for the new light-emitting diode technology The tests performed on the new light-emitting technology were conducted in the following sequence: 1. Measurements of 6000 K and 4100 K light-emitting diodes in the integrating sphere 2. Absolute photometry Astro with 6000 K diodes 3. Absolute photometry Astro with 4100 K diodes 5.1 Measurements of diode luminaires in integrating sphere Table 14: Measurements of LED luminaires in integrating sphere Astro 6000 K 4100 K Test report L L L L L L Number of lamp Sample 1 Sample 3 Sample 5 Sample 6 Sample 4 Sample 2 Voltage input Input current Input power THD (V) 1.84% 1.77% 1.68% 1.74% 1.74% 1.77% THD (A) 9.13% 8.68% 8.18% 8.13% 8.70% 8.02% Lamp voltage Lamp current Lamp power Measured flux Luminous efficacy Chromaticity x Chromaticity y IRC CCT ( K) advanced lighting technologies 23

41 5.2 Absolute photometry Astro with 6000 K diodes The three samples of Astro LED luminaires with 60 watts and 6000 K were tested on the mirror photometer. The corresponding photometric files are S , S and S , and the results are provided in the following summary table: Table 15: Astro 6000 K diodes on goniophotometer Photometric file S S S Lens Flat Flat Flat Driver Electronic Electronic Electronic Input power (watts) Maximum intensity Maximum intensity position 65.0 H, 55.0 V 65.0 H, 55.0 V 67.5 H, 50.0 V Maximum intensity at 90 vertical Maximum intensity at 80 vertical Transversal distribution Type II Type II Type II Longitudinal distribution Short Short Short LCS classification B1 U1 G1 B1 U1 G1 B1 U1 G1 Lumens (downward), street side Lumens (downward), house side Total lumens (downward) Lumens (upward), street side Lumens (upward), house side Total lumens (upward) Total lumens of the luminaire advanced lighting technologies 24

42 5.3 Absolute photometry Astro with 4100 K diodes The three samples of Astro LED luminaires with 60 watts and 4100 K were tested on the mirror photometer. The corresponding photometric files are S , S and S , and the results are provided in the following summary table: Table 16: Astro with 4100 K diodes on goniophotometer Photometric file S S S Lens Flat Flat Flat Driver Electronic Electronic Electronic Input power (watts) Maximum intensity Maximum intensity position 65.0 H, 50.0 V 67.5 H, 50.0 V 65.0 H, 50.0 V Maximum intensity at 90 vertical Maximum intensity at 80 vertical Transversal distribution Type II Type II Type II Longitudinal distribution Short Short Short LCS classification B1 U1 G1 B1 U1 G1 B1 U1 G1 Lumens (downward), street side Lumens (downward), house side Total lumens (downward) Lumens (upward), street side Lumens (upward), house side Total lumens (upward) Total lumens of the luminaire advanced lighting technologies 25

43

44 6. Tests on the 5 th and 8 th Street Astro LED luminaires with 60 watts were installed on the 5 th and 8 th Street in Saint-Gédéon-de- Beauce th Street Fifth Street is a residential street with low pedestrian conflict and with a width of 44 feet, consisting of 10 feet of easement on each side of the two traffic lanes of 24 feet. The pavement is R3 type and made of asphalt. Lamp posts are mounted on only one side of the street and 9.5 feet from the edge line with a span of 9.7 feet (including an 8-foot console and a luminous centre of 1.7 feet from the luminaire). Luminaires are 0.2 feet forward leaning from the edge line and are installed at a mounting height of 30 feet with spacing that varies between 172 and 207 feet th Street Eighth Street is also a residential street with low pedestrian conflict and with a total width of 50 feet, that is, 13 feet of easement on each side of the two traffic lanes of 24 feet. The pavement is R3 type and made of asphalt. Lamp posts are mounted on only one side of the street and 13 feet from the edge line with a span of 9.7 feet (including the 8-foot console and a luminous centre of 1.7 feet from the luminaire). Luminaires are 3.3 feet from the edge line and are installed at a mounting height of 30 feet with spacing that varies between 103 and 200 feet. However, there are two lamp posts which are shown without luminaires. The reason why there were no luminaires on these lamp posts was not specified. The spacing between lamp posts is variable and is sometimes greater than 6 times the mounting height. In some cases, the spacing is 224 feet and 372 feet. We then conducted calculations for illuminance and luminance for these two distinct cases. The first case corresponds to what is currently in place: 9 lamp posts and 7 luminaires (2 lamp posts being without luminaires). The second case corresponds to the scenario where one luminaire would be added to each of the two lamp posts currently without one, bringing the total to 9 lamp posts and 9 luminaires. advanced lighting technologies 27

45 6.3 Illuminance calculations for conventional HPS cobra-style luminaire Relative photometries of the cobra-style luminaire (reference lamp and reference ballast) were used in the Lighting Analysts lighting software AGI32 to evaluate its performance on the 5 th and 8 th Street in terms of illuminance and luminance. The overall maintenance factor used in the calculations is 0.90 in order to take into account the commercial ballast factor. These lighting levels correspond to the initial levels which were recorded after a period of 100 hours of operation. Since the luminaires were installed on one side only, we created two luminance calculation grids, the first one for the right lane and the second for the left lane. In addition, statistical zones of 120 feet and 150 feet were analyzed. Given that the lamps were used, the luminaires were evaluated with a new lamp powered from reference ballast. The luminaires were cleaned before these tests were conducted on the mirror photometer. Table 17: Used HPS luminaire in clean condition and a new lamp on reference ballast TEST S R1 S R1 S R1 Nominal power lamp DSS L DHS L LUMCAT (Watts) Cobra-100HPS-#48- C Cobra-100HPS-#58- C Cobra-100HPS-#60- C DTL USS L UHS L UT L TLL When the new lamp is tested on reference ballast, the result of lumens is obtained. The luminaire produces on average, for the three luminaires, lumens, which is the average of lumens, lumens and lumens. Table 18: New HPS lamp on reference ballast Test Voltage % Voltage (Vca) Current (A) Power Ballast (Watts) Power Lamp (Watts) Lumens BF L C1 REF advanced lighting technologies 28

46 It is therefore possible to reconstruct a figure for lumens distribution for a used luminaire in clean condition with a new lamp and on reference ballast. Overall, this test makes it possible to create a new HPS luminaire with a new lamp and a unitary ballast factor. Following this, it also becomes possible to create an HPS luminaire but with a ballast factor of 0.9. New lamp reference ballast: lumens Luminaire output (BF = 1.0): lumens (Luminaire efficiency: 71.3 %) Lumens downward: lumens Lumens upward: 332 lumens Lumens on the house side toward the ground Lumens street side toward the ground lumens lumens Figure 16: Overall efficiency for an HPS luminaire in clean condition and new lamp reference ballast (ballast factor of 1) advanced lighting technologies 29

47 New lamp reference ballast: lumens Luminaire output (BF = 0.9): lumens (Luminaire efficiency: 64.2 %) Lumens downward: lumens Lumens upward: 299 lumens Lumens on the house side toward the ground lumens lumens Lumens street side toward the ground Figure 17: Overall efficiency of an HPS luminaire in clean condition and new lamp ballast factor of 0.9 advanced lighting technologies 30

48 6.4 Illuminance calculations with 4100 K LED luminaire Total lumens: lumens Lumens on the house side toward the ground Lumens street side toward the ground lumens lumens Figure 18: Lumens distribution for a 4100 K LED luminaire (65 watts total) 6.5 Illuminance calculations with a 4100 K LED luminaire on 5 th Street Illumination simulation calculations were completed with the help of the software Visual Roadway Lighting. The parameters are as follows: i) 30 feet of mounting height; ii) 24 feet of street width; advanced lighting technologies 31

49 iii) iv) 150 feet of spacing; The lamp post is 9.5 feet from the edge line. The cross-arm measures 8 feet and the luminous centre 1.7 feet. Consequently, the span is 0.2 feet ( = 0.2 feet). The following three photometric files were selected: S , S and S The results are summarized in Table 19. It can be noted that on average for the three luminaires, the average luminance for the right lane is 0.31 cd/m 2 while for the left lane it is 0.45 cd/m 2. In this way, the necessary minimum of 0.3 cd/m 2 is surpassed. Lastly, it is interesting to note that good uniformity was obtained, i.e. maximum over minimum ratios and also the average over minimum ratio with the LED luminaire. In Table 20, the spacing between the poles was increased to 170 and 190 feet. advanced lighting technologies 32

50 Table 19: Simulation results for 5 th Street with 150 feet of spacing between luminaires and 4100 K LED luminaire S ies S ies S ies IES RP feet 150 feet 150 feet Average Luminance right side > 0.3 Average (cd/m 2 ) Maximum (cd/m 2 ) Minimum (cd/m 2 ) < 6 Average / minimum < 10 Maximum / minimum < 0.4 Veiling luminance ratio Luminance left side > 0.3 Average (cd/m 2 ) Maximum (cd/m 2 ) Minimum (cd/m 2 ) < 6 Average / minimum < 10 Maximum / minimum < 0.4 Veiling luminance ratio Average luminance entire street Average (cd/m 2 ) S ies S ies S ies IES RP feet 150 feet 150 feet Illuminance right side > 4 Average (lux) Maximum (lux) Minimum (lux) < 6 Average / minimum < 10 Maximum / minimum Illuminance left side > 4 Average (lux) Maximum (lux) Minimum (lux) < 6 Average / minimum < 10 Maximum / minimum Average illuminance entire street Average (lux) advanced lighting technologies 33

51 Table 20: Simulation results for 5 th Street with 170 feet and 190 feet of spacing between 4100 K LED luminaires S ies S ies IES RP feet 190 feet Luminance right side > 0.3 Average (cd/m 2 ) Maximum (cd/m 2 ) Minimum (cd/m 2 ) < 6 Average / minimum < 10 Maximum / minimum < 0.4 Veiling luminance ratio Luminance left side > 0.3 Average (cd/m 2 ) Maximum (cd/m 2 ) Minimum (cd/m 2 ) < 6 Average / minimum < 10 Maximum / minimum < 0.4 Veiling luminance ratio Average luminance entire street Average (cd/m 2 ) S ies S ies IES RP feet 190 feet Illuminance right side > 4 Average (lux) Maximum (lux) Minimum (lux) < 6 Average / minimum < 10 Maximum / minimum Illuminance left side > 4 Average (lux) Maximum (lux) Minimum (lux) < 6 Average / minimum < 10 Maximum / minimum Average illuminance entire street Average (lux) advanced lighting technologies 34

52 6.6 Illuminance calculations with HPS luminaire on 5 th Street Illuminance simulation calculations were completed using the software Visual Roadway Lighting. The parameters were as follows: v) 30 feet of mounting height; vi) 24 feet of street width; vii) 150 feet of spacing; viii) The lamp post is 9.5 feet from the edge line. The cross-arm is 8 feet and the luminous centre is 1.7 feet. Consequently, the span is 0.2 feet ( = 0.2 feet). The following three photometric files were selected: S , S and S It can be noted that the levels of luminance and illuminance are showing higher than those obtained with LED technology. However, the light source needlessly produces too much luminance for a street of residential type and with low traffic. LED technology is proving to be of interest as it reduces the required levels but still succeeds to reach an adequate level of uniformity. advanced lighting technologies 35

53 Table 21: Simulation results for 5 th Street with 150 feet of spacing between luminaires and HPS luminaire S ies S ies S ies IES RP feet 150 feet 150 feet Average Luminance right side > 0.3 Average (cd/m 2 ) Maximum (cd/m 2 ) Minimum (cd/m 2 ) < 6 Average / minimum < 10 Maximum / minimum < 0.4 Veiling luminance ratio Luminance left side > 0.3 Average (cd/m 2 ) Maximum (cd/m 2 ) Minimum (cd/m 2 ) < 6 Average / minimum < 10 Maximum / minimum < 0.4 Veiling luminance ratio Average luminance entire street Average (cd/m 2 ) S ies S ies S ies IES RP feet 150 feet 150 feet Illuminance right side > 4 Average (lux) Maximum (lux) Minimum (lux) < 6 Average / minimum < 10 Maximum / minimum Illuminance left side > 4 Average (lux) Maximum (lux) Minimum (lux) < 6 Average / minimum < 10 Maximum / minimum Average illuminance entire street Average (lux) advanced lighting technologies 36

54 7. Mesopic correction for LED luminaire Photometry is a measurement of emitted radiant energy weighted by the sensitivity of the human eye. At photopic levels, the levels of luminous efficacy are weighted by the function V(λ); however, at night vision, scotopic correction V (λ) is applied. For mesopic region, illuminance levels are between cd/m 2 and 10 cd/m 2, that is, between day vision and night vision. According to the University of Helsinki, V(λ) correction currently used to evaluate the quantity of lumens is only applied to conditions for which function V(λ) was obtained: It is acknowledged in CIE publication N 41 (Light as a true visual quantity: principles of measurement, 1978) that: Since the luminous efficiency function of the human eye is known to vary with a wide variety of viewing conditions, the assessment of radiant power can give accurate values only when the measured light corresponds to conditions under which V(λ) was obtained. Where do we need mesopic photometry? The most relevant mesopic lighting applications are street and road lighting and other outdoor lighting. The International Commission on Illumination set up a technical committee CIE TC Visual Performance in the Mesopic Range, which was involved in MOVE - Mesopic Optimisation of Visual Efficiency. In September 2010, the CIE published the committee s work on a photometry system based on mesopic photometry. 5 In North America, the ASSIST model outlines mesopic correction. LED luminaires were evaluated by taking into account both scotopic correction and photopic correction for three 4100 K samples, the tests L C1, L C1 and L C1. 5 Recommended System for Mesopic Photometry Based on Visual Performance, International Commission on Illumination / 01-Sep-2010 / 81 pages ISBN: advanced lighting technologies 37

55 Spectral Power Distribution Photopic (V?) Scotopic (V'?) SPD sample 1 SPD Sample 2 SPD Sample w/nm Wavelength (nm) Figure 19: Spectral power distribution of LED illuminance for the pilot project advanced lighting technologies 38

56 The following equation represents luminous flux in lumens for day vision: 780 nm 686 V ( ) P( ) d 380 nm where: P(λ) spectral power density in W/nm V(λ) photopic correction for day vision ø luminous flux in lumens for day vision The following equation represents luminous flux in lumens for night vision: 780 nm 1699 V '( ) P( ) d 380 nm where: P(λ) spectral power density in W/nm V (λ) scotopic correction for night vision ø luminous flux in lumens for night vision It is however important to note that the lumens provided by manufacturers always represent lumens for day vision with photopic correction. Table 11 presents experiment measurements. Table 22: Lumens measured in sphere, LDI LED 4100 K luminaires TEST Voltage (Vca) Current (A) Power (Watts) Photopic lumens (P) Scotopic lumens (S) RATIO S/P L C L C L C advanced lighting technologies 39

57 Luminous flux (lumens) Table 23: Photopic and scotopic lumens by wavelength, LDI LED 4100 K luminaire Day vision Night vision Photopic Scotopic Violet Lumens ( nm) 1 42 Blue Lumens ( nm) Green Lumens ( nm) Yellow Lumens ( nm) Orange Lumens ( nm) Red Lumens ( nm) Dark Red Lumens ( nm) 0 0 TOTAL lumens Photopic Scotopic Violet Lumens ( nm) Blue Lumens ( nm) Green Lumens ( nm) Yellow Lumens Orange Lumens ( nm) ( nm) Wavelength in nm Red Lumens ( nm) Dark Red Lumens ( nm) Figure 20: Distribution of luminous flux by type of vision LED 4100 K advanced lighting technologies 40

58 In summary, we can observe advantages of LED lighting for low levels of illuminance (typically road lighting and outdoor areas) based on spectral power distribution. A 2008 study entitled LED Street Lighting 6 states that: However, a lumen for lumen replacement scenario for LED outdoor retrofits does not account for improvements in color rendering, lighting distribution, and enhanced night time lighting conditions (Scotopic or mesopic vision advantages) that might allow for a reduction in total output from LED light sources relative to HPS. Recognizing the increasing interest in nighttime performance of LEDs, the DOE study notes that more energy savings would be possible if these factors were taken into account. 16 Because this is increasingly a part of the lighting design and energy planning discussion, evaluation of photopic and scotopic illuminance to characterize nighttime lighting performance of LED street light is included in this assessment. Therefore, traditional methods which rely on the quantity of emitted light based on day vision do not take into account spectral distribution of LEDs for night vision. This is the reason why traditional HPS technology should not be compared in terms of lumens. Given this scientific premise, detailed results of colorimetric measurements in sphere are listed in Annex 1. Table 24: Summary of measurements in integrating sphere, LED 4100 K luminaire Test # L C1 L C1 L C1 Type of light source 60W LED 4100 K 60W LED 4100 K 50W LED 4100 K Correlated Colour Temperature (CCT) K Colour Rendering Index (CRI) Chromaticity (x) Chromaticity (y) Lamp power (watts) Photopic lumens Scotopic lumens Photopic lumens /W Scotopic lumens /W Ratio (S/P) Scotopic to Photopic Lumens LED Street Lighting; Host Site: City of San Francisco, California; Final Report prepared in support of the U.S. DOE Solid-State Lighting Technology Demonstration Gateway Program and PG&E Emerging Technologies Program, December 2008; page 3 advanced lighting technologies 41

59 8. Mesopic correction for HPS luminaire A section of the present study conducts mesopic correction for LED lighting. The same approach was applied to HPS luminaire. The HPS luminaire was installed inside the integrating sphere in order to obtain spectral power density of the luminaire with reference to wavelength. To perform this exercise, the luminaire was installed inside the sphere and photometric measurements were taken. Figure 21: HPS luminaire installed inside the sphere Three luminaires were used, and in each case, reference ballast powered the lamp. This explains why the measured power of the lamp was very close to 100 watts, or approximately 99 watts. Table 25: Summary of measurements in integrating sphere for HPS luminaires advanced lighting technologies 42

60 Test # L C1 L C1 L C1 Type of light source HPS 100 watts Reference ballast HPS 100 watts Reference ballast HPS 100 watts Reference ballast Correlated Colour Temperature (CCT) K Colour Rendering Index (CRI) Chromaticity (x) Chromaticity (y) Lamp power (watts) advanced lighting technologies 43

61 Spectral Power Distribution Photopic (V?) Scotopic (V'?) SPD Sample 1 SPD Sample 2 SPD Sample w/nm Wavelength (nm) Figure 22: Spectral power distribution for HPS luminaire advanced lighting technologies 44

62 Figure 23: Spectral distribution for HPS luminaire test L C1 Table 32 lists photopic lumens and scotopic lumens. It should, however, be noted that nomenclature is not exact, given that a lumen is normally calculated with photopic correction. Table 26: Lumens measured in integrating sphere, HPS luminaire / reference ballast Test Lamp power (watts) Photopic lumens (P) Scotopic lumens (S) RATIO S/P L C L C L C advanced lighting technologies 45

63 Luminous flux (lumens) Table 27: Photopic and scotopic lumens according to wavelength, HPS luminaire Day vision Night vision Photopic Scotopic Violet Lumens ( nm) 1 29 Blue Lumens ( nm) Green Lumens ( nm) Yellow Lumens ( nm) Orange Lumens ( nm) Red Lumens ( nm) Dark Red Lumens ( nm) 0 0 TOTAL lumens Photopic Scotopic Violet Lumens ( nm) Blue Lumens Green Lumens ( nm) ( nm) Yellow Lumens ( nm) Wavelength in nm Orange Lumens ( nm) Red Lumens ( nm) Dark Red Lumens ( nm) Figure 24: Distribution of luminous flux by vision type - HPS advanced lighting technologies 46

64 9. ASSIST and LED versus HPS ASSIST model of the Lighting Research Center (LRC) lists the following values for an S/P ratio of 0.61 as well as Table 28: LED / HPS ratio by luminance level Photopic HPS Luminance ratio LED Luminance ratio LED / HPS ratio luminance S/P =0.61 M/P S/P = 1.52 M/P (A) (B) E = (B/A) (D) F = (D/A) G = F / E The data are represented in the following graph. It can be noted, for example, that for luminance of 0.2 cd/m 2, the ratio L mesopic / L photopic becomes 1.2 in the case for LED versus 0.8 for HPS. By dividing 1.2 by 0.8, we get a value of 1.5 (1.2 / 0.8 = 1.5). This is then the case of a pseudo increase by a factor of 1.5 thanks to LED distribution. It can be expressed as follows: y = x x x with a coefficient R 2 of 0.99 and x being photopic luminance. Therefore, for x =0.2 cd/m 2, we get a mesopic luminance of 1.49 cd/m 2 in the following equation: y = (0.2) (0.2) (0.2) = 1.49 advanced lighting technologies 47

65 Mesopic luminance LED / Mesopic luminance HPS Luminance mésopique / Luminance photopique HPS LED / 0.8 = 1.5 Increase of 50% in efficiency with LED versus HPS 0.2 cd/m 2 x 1.5 = 0.3 cd/m Photopic luminance (cd/m 2 ) Figure 25: Mesopic luminance / photopic luminance 2.5 LED / HPS 2.0 Polynomial (LED / HPS) y = x x x R 2 = / 0.8 = 1.5 Increase of 50% in efficiency with LED versus HPS 0.2 cd/m 2 x 1.5 = 0.3 cd/m Photopic luminance (cd/m 2 ) Figure 26: LED mesopic luminance / HPS mesopic luminance advanced lighting technologies 48

66 10. ASTRO 6000 K versus ASTRO 4100 K The temperature of a light source characterizes the tinge of the white colour. A light source is said to be cool if its correlated colour temperature value is high; it is said to be warm if its correlated colour temperature is low. High correlated colour temperature values represent bluish tinges, while low values represent yellowish tinges. Temperature colours and values CIE chromaticity diagram Figure 27: Comparison of colour temperatures of LED luminaires On the diagram CIE 1931 above, pure colours (monochromatic radiations) are located along the boundary curve, while mixed colours are inside the diagram and the white colour is in the centre. The straight line which links the two extreme points of the spectrum is called the purple line. advanced lighting technologies 49

67

68 11. Street luminaire efficiency Despite numerous structural improvements in road lighting made during the last decades, the traditional high intensity discharge technology does not answer the calls for reduction in light pollution and energy waste. The image below illustrates a number of problems associated with the use of conventional cobrastyle luminaires in the road network and the importance of optical systems in terms of the ability to control maximum intensity position, light pollution, nuisance and spill light. Figure 28: Spill light Efficiency of a luminaire is the ratio of luminous flux exiting the luminaire to total luminous flux emitted by the lamp. Indeed, no luminaire can reproduce 100% of the light emitted by the lamps. A significant share of this light is absorbed by different elements of the luminaire and is transformed into heat. Any other share of the light emitted into the atmosphere would be considered a loss, as it does not contribute to the lighting on the ground. advanced lighting technologies 51

69 Downward efficiency Upward flux fraction Figure 29: Downward efficiency and total efficiency Total efficiency of a luminaire is the ratio between luminous flux emitted by the luminaire and luminous flux of the lamps. Total efficiency comprises a downward component and an upward component. Downward efficiency is the ratio of luminous flux directed downward by the luminaire to total luminous flux of the lamp. It represents light lost through absorption by surfaces and through multiple reflection due to the construction of the device and its optical system. The upward flux fraction (UFF) is the ratio of luminous flux emitted in the upward direction to the total flux exiting the luminaire (upward + downward). This concept describes the capacity of a luminaire to control light pollution. advanced lighting technologies 52

70 12. Conclusion This study was able to validate LED technology. In contrast to traditional cobra-style, high pressure pressure luminaires of 100 watts which consume 130 watts, Astro luminaire consumes 65 watts with a superior efficiency of cd/m 2 per watt. Indeed, from figure S-1, it can be observed that the LED spectral power distribution represents an advantage for low-level road lighting (mesopic correction), which is used for residential lighting. Even if we do not take into account mesopic correction of LED, which is a major improvement for LED technology, the required levels of luminance (0.3 cd/m 2 ) are achieved. Table 29: Summary of luminance by side of the lane and technology (including LED mesopic correction) Power Right side Left side (watts) cd/m 2 cd/m 2 LED HPS We can then calculate efficiency in terms of cd/m 2 per watt by dividing average luminance by power consumption. Table 30: Efficiency by technology in cd/m 2 per electric watt Power Efficiency right side Efficiency left side (watts) cd/m 2 per watt cd/m 2 per watt LED HPS Following this approach, we can obtain the efficiency ratio between HPS and LED. Table 31: Gain in efficiency resulting from the replacement of HPS by LED Ratio right side Ratio left side RATIO LED / HPS By calculating the average for the two lanes, we obtain an increase in efficiency by 1.5, or by 50%. The annual energy savings, in monetary terms, for one street luminaire become $23.69 for a decrease in power consumption of 65 watts: from 130 watts to 65 watts. advanced lighting technologies 53

71 A comparative study shows that an Astro luminaire with a flat lens uses electrical energy better than a cobra head with a prismatic lens. Average efficiency in cd/m 2 per watt of Astro luminaire is 1.5 times greater than the average efficiency of a cobra head with a prismatic lens for a two-lane street. Moreover, the risk of accident due to the glare is greatly reduced with an Astro luminaire due to its absolute cutoff. Road lighting plays a key role in the management of public space. The growing tendency for cities and municipalities to create a strong local identity is naturally contributing to its management. Within this context, the perceived quality of lighting and the lamp post design are both considered to be crucial in the opinion of the citizens. Instant start lighting, low energy consumption, potential long life span of light-emitting diodes, reduction in costs for frequent lamp replacement, the absence of mercury and other polluting materials, the possibility to manage lighting via control systems are all arguments which highlight the benefit of using this new technology in road lighting. To this we can add the will of the local communities and companies to considerably reduce CO 2 emissions associated with lighting. advanced lighting technologies 54

72 Annex A: Sphere tests for 4100 K LED luminaire (L C1) advanced lighting technologies 55

73 advanced lighting technologies 56

74 advanced lighting technologies 57

75 advanced lighting technologies 58

76 advanced lighting technologies 59

77 Annex B: Photometer tests for 4100 K LED luminaire (S R1) advanced lighting technologies 60

78 advanced lighting technologies 61

79 advanced lighting technologies 62

80 advanced lighting technologies 63

81 advanced lighting technologies 64

82 advanced lighting technologies 65

83 advanced lighting technologies 66

84 advanced lighting technologies 67

85 advanced lighting technologies 68

86 advanced lighting technologies 69

87 advanced lighting technologies 70

88 advanced lighting technologies 71

89 Annex C: Photographs of the municipality of Saint-Gédéonde-Beauce Figure C-1: LED-type lighting, LDI company advanced lighting technologies 72

90 Figure C-2: LED-type lighting, LDI company advanced lighting technologies 73

91 Figure C-3: LED-type lighting, LDI company advanced lighting technologies 74

92 Figure C-4: LED-type lighting, LDI company advanced lighting technologies 75

93 Figure C-5: LED-type lighting by LDI company, as well as HPS lighting advanced lighting technologies 76

94 Figure C-6: LED-type lighting by LDI company, as well as HPS lighting advanced lighting technologies 77

95

96 Annex D: Required roadway lighting levels by the City of Ottawa for urban area 7 ROADWAY CLASSIFICATION AREA LUMINANCE GLARE ILLUMINANCE CLASSIFICATION Average Uniformity Veiling luminance luminance L v (Cd/m 2 ) ratio L v / L min ratio L Vmax / L v ARTERIAL Mixed use centre / Central area MAJOR COLLECTOR AND COLLECTOR ROADWAY CLASSIFICATION Employment / Enterprise area General urban area / Other Minimum maintained average E v (Lux) Uniformity ratio E v / E min Mixed use centre Employment / Enterprise area General urban area / Other AREA CLASSIFICATION LUMINANCE GLARE ILLUMINANCE COLLECTOR Mixed use centre / Central area Employment / Enterprise area General urban area / Other LOCAL Mixed use centre / Central area Employment / Enterprise area General urban area / Other Average luminance L v (Cd/m 2 ) Uniformity ratio L v / L min Veiling luminance ratio L Vmax / L v Minimum maintained average E v (Lux) Uniformity ratio E v / E min advanced lighting technologies 79

97

98 Annex E: Survey distributed to the municipality Survey results 1- Have you noticed that the street lights have been replaced? 4 respondents answered no 79 respondents answered yes 2- How did you discover that there was a new lighting system? 15 respondents answered by car 1 respondent answered by bicycle 18 respondents answered by foot 45 respondents answered from a home 3- Do you feel that the new, white lighting system has improved your visibility as pedestrian compared to the yellow-light, high pressure pressure (HPS) lighting system which was in place before? 28 respondents answered yes 29 respondents answered no 16 respondents answered approximately the same 4- Do you feel that the new, white lighting system has improved your visibility as driver? 25 respondents answered yes 34 respondents answered no 14 respondents answered approximately the same 5- According to you, the new street lighting system makes the road: Safe: 25 yes, 19 no, 15 same. More pleasant: 28 yes, 15 no, 11 same. Too bright: 4 yes, 35 no, 9 same. Too dark: 28 yes, 29 no, 7 same. Better colour distinction: 19 yes, 20 no, 10 same. 6- Overall, are you satisfied with the new street lighting system? 41 respondents answered yes 26 respondents answered no 9 respondents answered approximately the same advanced lighting technologies 81

99

100 State State Annex F: Site measurements of the operating state Date: July 8, Hour of the day July 9, Hour of the day Figure F-1: Operating state of an LED luminaire measured in summer advanced lighting technologies 83

101 State State January 3, Hour of the day January 2, , Hour of the day Figure F-2: Operating state of an LED luminaire measured in winter advanced lighting technologies 84

102 Annex G: Site measurements of LED luminaire lighting Table G-1: Site measurements, June 21, 2010 Part 1 (east side pole 7) Lux Lux Lux Lux Distance A (3 FEET) B (9 FEET) C (15 FEET) D (21 FEET) Part 2 (west side pole 7) Lux Lux Lux Lux Distance A (3 FEET) B (9 FEET) C (15 FEET) D (21 FEET) advanced lighting technologies 85

103

104 Annex H: Energy savings This annex was prepared as additional information to the main report in order to quantify the potential energy savings. Concerning the calculations of energy savings, it should be noted that the luminaires are part of an electrical network that is not measured by a meter. Admittedly, energy is not measured on the majority of street luminaires. For such cases, the following billing procedure was put in place. According to the document entitled Distributor s Rates and Conditions, in chapter 9, rates are fixed in the following manner in order to determine consumption within the framework of general public lighting service: Rate The rate for general public lighting service is 8.82 per kilowatthour for the supply of electricity. 9.5 Determination of consumption As a rule, the energy consumption is not metered. However, the Distributor may meter the consumption if it deems appropriate. When it is not metered, the energy consumption is the product of the connected load and 345 hours of monthly utilization. In the case of tunnels or other facilities that remain lighted 24 hours a day, the energy consumption is the product of the connected load and 720 hours of monthly utilization. To establish the connected load, the Distributor takes into account the rated power of the bulb and accessories. By adopting the following hypotheses: 1) Conventional, HPS, 100 watt lamp: 130 watts including ballast loss 2) LED lighting: 55 watts 3) 345 hours per monthly utilization 4) Energy cost: 8.82 / kwh Power savings are then = 75 watts Annual consumption is then: advanced lighting technologies 87

105 75 watts x 345 hours / month x 12 month / year = Wh per year or 310 kwh Annual monetary savings per luminaire are: 310 kwh x $ / kwh = $27.34 In the case where a LED luminaire consumes 65 watts and an HPS luminaire 130 watts, we obtain 65 watts in power savings (130 watts 65 watts). It is possible to take the monetary savings of $27.34 that we get when power consumption is reduced to 75 watts, and to cross-multiply to get monetary savings for a reduction of 65 watts. $27.34 x (65/75) = $23.69 Annual monetary savings in energy for one street luminaire then become $23.69 for a reduction of power of 65 watts, from 130 watts to 65 watts. advanced lighting technologies 88

106 Annex I: Specifications of the City of Los Angeles This annex provides specifications for LED luminaires provided by the municipality of Los Angeles, which undertook a major street-luminaire replacement project in residential sectors. It is interesting to note that the following specifications were included: i) Type II and Type III luminaires; ii) power factor greater than 0.9; iii) a nominal CCT of 4000 K; 8 iv) a minimum IP of IP66. 8 Personal communication, March 2011, Ed Ebrahimian (City of Los Angeles) advanced lighting technologies 89

107 advanced lighting technologies 90

108 advanced lighting technologies 91