An efficient LED lamp that illuminates the street with high quality is presented. The luminaire shows high optical efficiency, high optical utilization factor, low glare, and illuminates the street with high uniformity. The concept is simple but effective: a cluster of LEDs with TIR lenses are put inside a reflective box, which is covered with a microlens sheet; the reflective cavity improves efficiency by light recycling; each TIR lens collimates the LED light for the microlens array; and the microlens sheet uniformly distributes light only into the street. We verify its feasibility by Monte Carlo ray-tracing for the main types of road lighting arrangements: central, zigzag, and single-side pole positions.
©2013 Optical Society of America
In the past few years, developing energy-efficient street-lighting with light-emitting diodes (LEDs) has gained an enormous interest. LED luminaires also have the potential of increasing illumination uniformity and glare reduction, which improves both the eye comfort and the visual discrimination ability of car drivers [1–4]. Traditional street lighting technologies, such as high-pressure sodium or mercury, emit light in all directions, and consequently the light distribution is difficult to control. This is why a common street luminaire usually has defects such as glare, non-uniform light pattern, upward reflected light, light pollution, and waste of energy, as shown in Fig. 1(a). In this context, LED luminaires have the potential to deliver precise light patterns to maximize illumination performance by directing light to the appropriate area. Because of the small size and directional light emission of LEDs, non-imaging optics may reshape the radiation pattern of LEDs to distribute their light only where it is needed. Figures 1(b) and 1(c) show a schematic comparison. In Fig. 1(b), nearly half of the light will be lost out of the road, the illumination is not uniform, and the light falling outside may cause eye discomfort for drivers and pedestrians. In the other side, using LEDs with appropriate design, energy is saved because it is not wasted to the sky or the sides of the street, as shown in Fig. 1(c). In principle, a LED street-light with an appropriate design may provide a rectangular light pattern that just cover the roadway, without glare light, without light creeping into yards, and without shining into the sky.
In this paper, we propose a novel scheme of LED street lighting. In Section 2, we present the optical concept. Then, in Section 3, we follow with some preliminary issues before the optical analysis. In Section 4, we analyze its optical performance by assessing the illumination uniformity and optical efficiency. And we give our conclusions in Sec. 5.
2. New concept of LED luminaire
LED light can be shaped and projected into the street by means of different optical methods. Recent research in LED luminaires focuses on reflective and refractive non-imaging optics; which may be free-form reflectors, free-form lenses, or reflective-refractive combinations [5–12]. Although the approaches reported are very good, the ultimate performance has not been met. Several challenges remain like the optical design complexity, manufacturing difficulty, and tolerancing.
We propose a high-performance concept of LED street lighting, as shown in Fig. 2, which addresses these problems. The lighting module is easy to design and fabricate because it only needs a cluster of LEDs, total internal reflection (TIR) lenses, one reflective cavity, and one microlens array plate. The LEDs are placed in a reflecting cavity of high reflectance to recycle light and increase the optical efficiency. Each LED is covered with a TIR lens that has a function with narrowing divergence angle. TIR lens collimates the LED light as much as possible to maximize the beam shaping capabilities of the microlens array . The microlens plate plays a key role in the whole module because it is both beam shaper and homogenizer, and increases tolerancing. The result is that the street lamp distributes the light within the street surface with high performance; showing low glare, low light pollution, high optical efficiency, high optical utilization factor, and high illumination uniformity.
3. Preliminaries to optical analysis
Here we review some important issues before Section 4. We describe the pole configurations, we describe and model every optical element of the lighting module, and we specify the desired lighting performance.
3.1 Types of street
The light distribution characteristics depend on the type of roadway. There are many types of streets and placement of poles. Figure 3(a) shows the three most popular pole placements: central, zigzag, and single-sided [14,15]. In this paper, we analyze the LED luminaire with these street lighting types. The central type is mainly used in tunnel lighting, and its optical implementation is the simplest. Zigzag and single-sided types are more frequently used in general street lighting, but its optical implementation is more difficult because the luminaire must emit light in a quite asymmetric radiation pattern .
The light distribution not only depends on the pole placement, it also depends on the lamp mounting height, pole pitch, road width, arm length, and angle of tilt [14–16]. Here we set these parameters to those shown in Figs. 3(a) and 3(b). The width of roadway is 12 m (maybe for 4 lines), the pole pitch is 20 m (initial value for analysis), and the mounting height of street light is 10 m. Module tilted angle (ϕ) is 0° for the central type, and 15° for zigzag and single-sided types, as shown in Fig. 3(b). Arm length (s) are 6 m for the central type, 2.61 m for zigzag type, and 3 m for single-sided type, respectively.
3.2 Microlens array
The microlens plate must have the ability to confine LED light within the street boundaries. A structured microlens array has the beam shaping capabilities of diffractive elements plus the homogenization properties of random diffusers [17,18]; and more important, it operates efficiently under the broadband illumination of LEDs, which is not possible with diffractive plates. However, the structure must be small for achieving light homogenization, and then the fabrication process usually is expensive and complex.
We recently demonstrated and developed a high-performance, large-area, and low-cost structured microlens array . For street lighting, such microlenses may produce a rectangular light pattern that just cover the roadway without wasting light out of this region. The microlens array must have an asymmetrical lens structure so that different light distributions result along X and Y street directions, as shown in Fig. 4(a). After analysis and ray-trace simulations with perfect collimated light, tracing each time 20 million rays, we obtained the basic structure of microlenses [17,18]. Figure 4(a) shows the microscopic structure of the microlens plate used in our luminaire. Figure 4(b) shows the light pattern that is obtained if the dimensions of each microlens are 8 mm (X), 4.4 mm (Y), and 4.7 mm (Z). The one-shot transmittance of the microlens plate (acrylic, n = 1.49) is 90.4% for a collimated beam of light. The light pattern is practically a rectangle, which is the perfect shape for street lighting. The size of light pattern at 10 m distance is 22 m × 12 m.
To collimate LED light we attach a TIR lens, but collimation is not perfect. And then, when using LEDs + TIR lens, the light pattern should spread a little. This dispersed light is useful in our analysis for cover the practical area of 30 m × 14 m, which includes the light falling 1 m out of street for pedestrian use. A key feature is that each microlens distributes the light in the entire illumination pattern, and then manufacturing tolerances may be very high.
3.3 TIR lens
The design and modeling of TIR lens depends on how good the LED optical model is [20–23]. To build up the LED model, first an LED must be chosen. We selected a modern high-power type, the Cree XP-E warm-white LED. The midfield method  was implemented to build up the LED model, which tests and compares with experiment the angular intensity distribution in the midfield region. The model was precisely built, getting normalized cross correlations higher than 99.5% for the region of interest . Once the LED model is precise we proceed to design the TIR lens. The designed lens should have a collimating effect on LED to enhance the performance of microlens plate. The LED emission angle should be narrowed to approach collimated light. The light of Cree XP-E LED has a full width at half maximum (FWHM) of 120°. Therefore, based on the selected LED, we designed and modeled the TIR lens shown in Fig. 5(a). The refractive index of lens is 1.59 (polycarbonate), and its dimensions are shown in Fig. 5(a). The TIR lens narrows the FWHM divergence angle from 120° to 10°, and its optical efficiency is 92.8%.
3.4 Reflecting cavity
Although high-power LEDs produce many lumens per unit, several LEDs must be mounted on the street light panel to reach the luminous requirements. And then a housing box is needed to support and protect LEDs from environment. In our design, the housing is also a reflecting cavity with the property of photon recycling for increasing the optical efficiency of the whole module . We modeled the inner walls of the cavity as high reflective sheet, with a reflectance of 95%. The inner surfaces included 4 side walls and 1 bottom, as shown in Fig. 5(b). We used two types of cavity: for central and zigzag setups we modeled the box shown in the upper of Fig. 5(b); for single-sided type we proposed and modeled a new type of cavity, which is shown in the bottom of Fig. 5(b). In order to balance LED-to-LED separation, volume and weight, in our analysis we set the cavity size as: L = 360 mm, W = 280 mm, H = 50.75 mm, H1 = 88.26 mm, H2 = 13.24 mm. For reference, in Fig. 5(b), length L is along the driving direction. Right subfigures in 5(b) show the placement of LEDs.
3.5. Performance specifications
In order to assess the lighting performance, the luminaire should achieve some requirements. Table 1 shows these specifications. In power consumption we set an upper limit of 120 W per lighting module, a low value in the LED street lighting market. Illuminance levels and uniformity are set higher than the highest standards specified by CIE and Illuminating Engineering Society of North America (IESNA) . Central illuminance (Ec) is that in front of lamp, at the center of the street. The uniformity is calculated by the ratio of minimum illuminance (Emin.) to the average illuminance (Eavg.).
To achieve specifications of Table 1, it is important to define the target area for assessment. Using perfectly collimated light, the illuminated area should be 22 m × 12 m, and then when using LEDs the light distributes around this region. A single module mainly illuminates that area, intense light falls only there, and then glare is minimized [14–16]. Glare is very low because pedestrian or drivers receive very low veiling light in their field of view, which is a narrow cone parallel to the driving direction.
To reduce power consumption, the optical efficiency of the whole system should be high. For assessing the efficiency of light distribution we use the delivery factor (DF). The DF of one module is the ratio of the light flux on the target region of street to that emitted by the module. For assessing the total optical efficiency we calculate the optical utilization factor (OUF). The OUF is the product of DF and the optical efficiency of module. In other words, the OUF is the ratio of the light flux on the street-target-region to that emitted by all LEDs inside the module. We assessed these factors for two different illumination or target areas: small and practical area, as shown in Figs. 6(a) and 6(b). The small area is 20 m × 12 m, which is defined by the street itself and the pole separation. The practical area is 30 m × 14 m because all the light falling on the street and that falling 1 m out of street (for pedestrian usage) is useful.
For assessing both the average illuminance and the illumination uniformity we analyze the area in Fig. 6(c), which is 60 m × 12 m. This area is defined by a set of 3 modules, and then illumination uniformity assesses the light of three neighboring luminaries. According to the IESNA regulation for roadway illumination , the average illuminance and uniformity are calculated from a set of test points, as shown in Fig. 6(d). The points are distributed, in a roadway of width (WR), in lines at 1/8 WR distance from the sides of roadway, and the distance between neighboring lines is ¼ WR. In our study, WR is 12 m so that ¼ WR is 3 m and 1/8 WR is 1.5 m. The maximum spacing between neighboring measurement points must be 5 m along the direction of traffic flow, and in our calculations it was 2 m. Both the average illuminance and uniformity are computed from all these test points.
4. Optical analysis
In this section the performance of the luminaire is optically analyzed by Monte Carlo ray tracing. Once we have established the types of roadway, the specifications of street lighting, and the optical model of the whole module, we can begin the optical analysis. The lighting cavity contains 48 XP-E LEDs with TIR lenses, arranged as shown in Fig. 5(b). We set the LED light flux to 129.3 lm, a flux measured at 2.38W power, so that the whole module should emit 6155 lm. The module consumes a power of 114.24 W, which satisfies the first parameter of Table 1. A total of 48 million rays were traced to simulate the optical performance of each module, one million rays per LED. The module analyzed in Sec. 4.1 and 4.2 has an optical efficiency of 87.8%, and the module of Sec. 4.3 has an optical efficiency of 87.2%.
4.1 Central pole placement
The central pole arrangement is frequently implemented in tunnel lighting [14,15]. Each module is mounted at the central position of roadway. Figure 7 shows the illumination patterns from a single, and three luminaires. The illuminance distribution of a single light module is shown in Fig. 7(a). The module achieves a DF of 78.6% in the small area (20 m × 12 m), and 92.3% in the practical area (30 m × 14 m). The OUF is 69% in the small area, and it is 81% in the practical area. The central illuminance is 20 lux, and then the 2nd parameter of Table 1 is achieved.
Although the illumination uniformity is good at a pole pitch of 20 m, by changing pole spacing one may find a very homogenous light pattern, as shown in Fig. 7(b). According to our simulation results, the best pole spacing is 21.5 m, which gives a high uniformity of 0.877. The average illuminance is 19.2 lux, and then the 3rd and 4th parameters of Table 1 are achieved.
4.2 Zigzag pole placement
The ZigZag arrangement is implemented in open roads. Each module is mounted at 2.61 m from alternating sides of roadway. Figure 8 shows the illumination distributions. The illuminance pattern of single street light module is shown in Fig. 8(a). The module achieves a DF of 73% in the small area, and 90% in the practical area. The OUF is 63.7% in the small area, and it is 79% in the practical area. The central illuminance is 17.5 lux, and then the 2nd parameter of Table 1 is achieved.
Since each module is placed near the side of street, it should be tilted to fill the street with light. The illumination pattern becomes trapezoidal shaped because the beam rotation. Fortunately, the light fits well the street if all trapezoid patterns arrange in Zigzag type. Figure 8(b) shows the light pattern due to three neighboring modules in a street area of 60 m × 12 m. The illumination uniformity is acceptable for a pole pitch of 20 m. By changing the luminarie spacing, one may find a more homogenous light pattern. According to our simulation results, the best pole spacing is 22.5 m, which gives a good uniformity of 0.633. The average illuminance is 17.2 lux, and then all parameters of Table 1 are achieved.
4.3 Single-sided pole placement
The single-sided configuration is essential in roads of high traffic flow, and is the most popular luminaire arrangement. The street is illuminated by a single line of modules placed at the side of street. Because each module must be tilted to fill the street with light, the illumination pattern has trapezoidal shape. This is a problem because the light pattern of neighboring modules should be rectangular shaped to fit together. This issue may be visualized as a mismatch between the light beam and the street surface, as shown in Fig. 9(a). Our solution was to shape the light beam with an asymmetric and opposite distortion by making an angle between the LEDs and the microlens, as shown in Fig. 9(b). The beam distortion compensates the street mismatch so that the light pattern on the street nearly has a rectangular shape. Figure 9(c) shows the illuminance pattern of single street light module. The module achieves a DF of 66.4% in the small area, and 84.4% in the practical area. The OUF is 58% in the small area, and it is 74% in the practical area. The central illuminance is 17.9 lux, and then the 2nd parameter of Table 1 is achieved.
Although the illumination uniformity is good for a pole pitch of 20 m, by changing pole spacing one may increase the homogeneity of light pattern, as shown in Fig. 9(d). According to our simulation results, the best pole spacing is 22 m, which gives a high uniformity of 0.806. The average illuminance is 16.2 lux, and then the 3rd and 4th parameters of Table 1 are achieved.
We have proposed a new LED street luminaire, which has been explained, and its performance analyzed. This LED lamp delivers a rectangle light pattern that maximizes illumination performance by directing light only where is needed. The light is efficiently and homogeneously distributed; which is a requirement for reducing glare, and improving both the eye comfort and the visual discrimination ability of car drivers.
The optical concept is simple and effective: an array of LEDs with TIR lenses are put inside a reflective cavity, which is covered with a microlens sheet; the reflective box improves efficiency by recycling the back-reflected light; each TIR lens efficiently collimates the LED light for the microlens array; and the microlens plate combines beam shaping and light diffusion to efficiently and homogeneously deliver light into the street.
We assessed the luminaire performance by Monte Carlo ray-tracing for the main types of street pole arrangements: central, zigzag, and single-side. The lamp achieved high optical utilization factors from 58% to 81%, which are high even compared with excellent designs, for example 45% achieved in . Moreover, in all cases, our LED street-light achieved the expected performance, specified by Table 1, which was set at the highest levels specified by the market and the IESNA. Based on this performance and the intrinsic tolerancing of microlens optics, we believe this design is the best ever reported.
This study was supported in part by the National Central University’s Plan to Develop First-class Universities and Top-level Research Centers Grants 995939 and 100G-903-2, and was also sponsored by the National Science Council of the Republic of China under contracts 99-2623-E-008-002-ET and NSC100-3113-E-008-001.
References and links
1. A. Haans and Y. A. W. de Kort, “Light distribution in dynamic street lighting: two experimental studies on its effects on perceived safety, prospect, concealment, and escape,” J. Environ. Psychol. 32(4), 342–352 (2012). [CrossRef]
2. K. Painter, “The influence of street lighting improvements on crime, fear, and pedestrian street use, after dark,” Landsc. Urban Plan. 35(2-3), 193–201 (1996). [CrossRef]
3. R. Müllner and A. Riener, “An energy efficient pedestrian aware smart street lighting system,” Int. J. Pervas. Comput. Commun. 7(2), 147–161 (2011) (IJPCC). [CrossRef]
4. C. C. Sun, C. J. Jiang, Y. C. Chen, and T. H. Yang, “Equivalent luminance for three types of street lamps based on white LEDs,” To be submitted.
7. J. W. Pan, S. H. Tu, W. S. Sun, C. M. Wang, and J. Y. Chang, “Integration of non-Lambertian LED and reflective optical element as efficient street lamp,” Opt. Express 18(S2Suppl 2), A221–A230 (2010). [CrossRef] [PubMed]
9. Y. C. Lo, K. T. Huang, X. H. Lee, and C. C. Sun, “Optical design of a Butterfly lens for a street light based on a double-cluster LED,” Microelectron. Reliab. 52(5), 889–893 (2012). [CrossRef]
10. C. H. Jen, Y. Y. Chen, A. J. W. Whang, and M. J. Lu, “Non-axisymmetrical freeform design for short LED street lamp,” Proc. SPIE 8123, 812307, 812307-7 (2011). [CrossRef]
11. J. Jiang, S. To, W. B. Lee, and B. Cheung, “Optical design of a freeform TIR lens for LED streetlight,” Optik (Stuttg.) 121(19), 1761–1765 (2010). [CrossRef]
12. C. C. Sun, W. T. Chien, I. Moreno, C. T. Hsieh, M. C. Lin, S. L. Hsiao, and X. H. Lee, “Calculating model of light transmission efficiency of diffusers attached to a lighting cavity,” Opt. Express 18(6), 6137–6148 (2010). [CrossRef] [PubMed]
13. X. H. Lee, C. S. Wu, K. H. Lee, T. H. Yang, and C. C. Sun, “An optical-adjustable illumination pattern with surface-structured diffuser compensated by index-matching liquid,” Opt. Laser Technol. 49, 153–155 (2013). [CrossRef]
14. Illuminating Engineering Society of North America, The IESNA Lighting Handbook: Reference and Application, 9th ed. (IESNA, 2000).
15. A. Pachamanov and D. Pachamanova, “Optimization of the light distribution of luminaries for tunnel and street lighting,” Eng. Optim. 40(1), 47–65 (2008). [CrossRef]
16. I. Moreno, “LED street lighting: modeling and design,” To be submitted.
17. T. R. M. Sales, “Structured microlens arrays for beam shaping,” Opt. Eng. 42(11), 3084–3085 (2003). [CrossRef]
18. X. H. Lee, J. L. Tsai, S. H. Ma, and C. C. Sun, “Surface-structured diffuser by iterative down-size molding with glass sintering technology,” Opt. Express 20(6), 6135–6145 (2012). [CrossRef] [PubMed]
19. V. N. Mahajan, Optical Imaging and Aberrations: Part I. Ray Geometrical Optics (SPIE Press, 1998).
20. A. Domhardt, S. Weingaertner, U. Rohlfing, and U. Lemmer, “TIR optics for non-rotationally symmetric illumination design,” Proc. SPIE 7103, 710304, 710304-11 (2008). [CrossRef]
21. J. J. Chen and C. T. Lin, “Freeform surface design for a light-emitting diode-based collimating lens,” Opt. Eng. 49(9), 093001 (2010). [CrossRef]
22. C. C. Sun, T. X. Lee, S. H. Ma, Y. L. Lee, and S. M. Huang, “Precise optical modeling for LED lighting verified by cross correlation in the midfield region,” Opt. Lett. 31(14), 2193–2195 (2006). [CrossRef] [PubMed]