PERFORMANCE EVALUATIONS OF LED HEADLAMPS

Transcription

1 International Journal of Automotive Technology, Vol. 11, No. 5, pp (2010) DOI /s Copyright 2010 KSAE /2010/ PERFORMANCE EVALUATIONS OF LED HEADLAMPS B. KANG 1)*, B. YONG 2) and K. PARK 3) 1) Graduate Schools, Kyunghee University, Seoul , Korea (Korea Automobile Testing & Research Institute, Hwaseong-si, Gyeonggi , Korea) 2) School of Mechanical & Automotive Engineering, Kyungil University, Gyeongbuk , Korea 3) College of Advanced Technology, Kyunghee University, Gyeonggi , Korea (Received 16 March 2009; Revised 6 February 2010) ABSTRACT Due to rapid progress in LED (Light Emitting Diode) technology, LEDs have become increasingly prevalent in applications to automotive lighting. This paper studies recent trends of LED headlamps along with various experiments regarding photometric characteristics, glare, and visibility. Experimental results are compared to those of halogen and HID (high intensity discharge) headlamps. LED headlamps are found to be still inferior to conventional headlamps with respect to glare. However, many properties of LED headlamps (luminous intensity, beam width, energy efficiency, and design) are competitive with halogen and HID headlamps. The findings in this investigation will henceforth be reported to the Korean Motor Vehicle Safety Standard (KMVSS). KEY WORDS : LED headlamp, Photometric performance, Glare, Spectral analysis, Cut-off line 1. INTRODUCTION *Corresponding author. headlamp@ts2020.kr In the past, automotive lighting equipment has been used simply for illumination and signaling. Electronic control technology for automotive lighting has developed remarkably over the past few decades; this technology is frequently applied to automotive headlamps such as AHLD (Automatic Headlamp Leveling Device), Xenon gas discharge lamps (High Intensity Discharge lamp, HID), and AFLS (Adaptive Front Lighting System). Accordingly, the front lighting system has evolved for optimized night driving. Further improvements to headlamp performance come from an innovative automotive lighting solution based on LED technology. The development of high power white LED (100 lm/w or higher) make sophisticated automobile headlamps possible. LED headlamps can deliver greatly improved front lighting functions and attractive designs (Dubosc, 2005; Scholl, 2005). Because design concepts and characteristics of LED headlamps are unlike those of conventional headlamps, much research has recently been carried out (Neumann et al., 2007; Apfelbeck, 2007). LED headlamps have many advantages compared to traditional headlamps. First, LED headlamps have longer lifetimes; they almost surpass the lifetime of automobiles. Second, improved LED technology allows solid-state reliability and nearly limitless styling potential of automotive headlamp architecture using combinations of LED modules (Kang et al., 2006). Accordingly, many automotive manufacturers and headlamp suppliers are moving into global markets backed by heavy investments and advanced technologies to capture the LED headlamp market of the next generation (Hamm, 2005). Third, LED headlamps produce less pollution and consume less energy than conventional headlamps. For example, a typical LED headlamp uses about 80% of the energy consumption of a halogen headlamp. While delivering high flux, however, LED light sources yield relatively low luminance, which results in completely different light patterns from halogen or HID headlamps. The enhancement of the light patterns and high luminous intensity are necessary to generate a clear cut-off line. Additionally, LED light sources are highly sensitive to temperature variations. Effective heat dissipations in the light sources and drive modules should be manifested. In this study, we review recent trends in international and domestic LED headlamp industries. Using selected headlamps (halogen, HID, and prototypes of LED headlamps), we evaluate photometric performance, glare, and sensitivity glare through simulation and experimentation. Finally, we compare the test results and draw conclusions. This investigation will be reported to establish the KMVSS (Korean Motor Vehicle Safety Standard) for automotive headlamps. 2. RECENT TRENDS IN LED HEADLAMPS For automotive headlamps, LEDs should fulfill light output requirements such as luminous efficiency, luminous flux, and heat dissipation. Because high-brightness white LEDs 737

2 738 B. KANG, B. YONG and K. PARK already outperform ordinary halogen lamps (20 lm/w) in luminous performance, they can successfully compete with existing headlamps on the market. Although photometrically adequate LED headlamps only recently became commercially available, LEDs have been developed in the U.S., Europe, and Japan for decades. Lumileds is a joint venture between HP and Philips that occupies a noticeable share in the market with its Luxeon high bright LEDs for automobiles. Lumileds is developing not only high-brightness LEDs but also high efficiency LEDs. Osram uses technology based on halogen and Xenon lamps to produce O-Star LEDs. In the 2003 GTB Congress, Stanley presented prototypes of LED headlamps and LED modules. This projection type LED headlamp, composed of 16 white LEDs, produced up to 490 lm of luminous flux and 23,500 cd of luminous intensity. Nichia shares patent rights for GaN LEDs with Osram and released an LED with high efficiency and high capacity of more than 100 lm/w in Some Korean manufacturers, such as Seoul Semi Conductor, Samsung Electrics, LG Innotech, Epivalley, and EpiPlus, have recently developed prototypes of highbrightness LEDs comparable to imported products. Samsung Electrics has introduced a blue colored LED, which is necessary for the realization of full color, but further development is still necessary to compete with the products of Nichia and Toyota Gosei. The major U.S. automobile manufacturers GM, Ford, and Chrysler utilize simple and sharp images of high performance projection type LED headlamps. Most Japanese automobile companies, such as Honda, Toyota, Nissan, Mazda, and Mitsubishi, provide multi-chip LED headlamps, which result in simple designs as well as improved photometric performance. European automobile manufacturers, such as Audi, Renault, Stola, and Volkswagen, employ crystal multi-chip LED headlamps that take into consideration mass productivity and refined simple design. Meanwhile, in Korea, LEDs are only used for RCLs (Rear Combination Lamps) and HMSLs (High Mounted Stop Lamps). With the introduction of high brightness LEDs, Korean headlamp manufacturers began developing LED headlamps in The products are still under development to resolve issues of heat dissipation, chromaticity, and light patterns, but it is expected that these deficiencies will soon be resolved (Kang et al., 2007). 3. PERFORMANCE EVALUATIONS To comply with the KMVSS, the performance and durability of the LED headlamps should equal or exceed those of halogen and/or HID lamps. Additionally, it is necessary for LED headlamps to satisfy related international safety regulations. Using the prototype LED headlamps shown in Figure 1, we carried out functional evaluations on photometric performance, glare, and visibility. Because halogen (H7 lamp type) and HID (D2R lamp type) headlamps (also depicted in Figure 1) are already available on the Korean Figure 1. Test samples. market, the performance of these conventional headlamps is compared to the prototypes of LED headlamps (Kang et al., 2008) Photometric Performance Luminous intensity and chromaticity The luminous intensity and chromaticity were evaluated in a dark room undisturbed by any other light sources. The voltages and currents used are shown in Table 1. The experimental arrangement of the photometric performance evaluation is depicted in Figure 2. The luminous intensity is measured from a distance of 25 m, and the chromaticity is observed at a distance of 4.25 m after the experimental apparatus has been set up in accordance to the visual aiming in Figure 3. The luminous intensities at specific points are illustrated in Table 2. LED headlamps show better performance than halogen and HID headlamps. However, they also yield Table 1. Input voltage and current. Voltage (V) Current (A) Halogen HID S-model LED M-model LED Figure 2. Schematics of experimental set-up.

3 PERFORMANCE EVALUATIONS OF LED HEADLAMPS 739 Figure 5. Dimensions for a 5 lx line simulation. Figure 3. Visual aiming. Table 2. Luminous intensity. Halogen HID S-model LED (unit: cd) M-model LED B50L R 15,189 14,101 14,850 21,637 50V 14,735 13,806 21,006 16,333 25R 6,446 12,650 12,802 12,027 25L 4,287 12,790 9,523 11,280 Zone III Max. 19,977 21,168 37,737 26,421 higher luminous intensity in Zone III (eye point area of subject vehicle s driver), which may result in discomfort glare. The chromaticity measurements are presented in Figure 4. The white light from LED headlamps falls in close proximity to the blue-green area. This finding implies that the prototype LED lamps do not satisfy the chromaticity standard of KMVSS, and thus, further development is necessary to avoid uncomfortable glare Simulation of iso road illumination The simulation for iso road illumination is carried out with a 5 lx line, the standard measurement level of road illumination. As shown in Figure 5, the simulation conditions of the actual test vehicle specify the distance between the two headlamps of the vehicle, the lamp height from the road surface, the vertical distance from the centerline of the vehicle to the eyes, the height from the ground to the eyes, and the distance from the center of the headlamp to the eyes. The simulation results for iso road illumination are presented in Figure 6. HID provides the highest performance with respect to illumination length (5 lx line) and maximum beam width, followed by LED and halogen headlamps. The actual vehicle test with respect to the 5 lx line is intended to measure the minimum possible distance for recognizing objects on the road. The experiments shown in Figure 7 were performed at the testing site of KATRI (Korea Automobile Testing & Research Institute). The road Figure 4. Chromaticity coordinates. Figure 6. Simulation results of a 5 lx line.

4 740 B. KANG, B. YONG and K. PARK Figure 9. Measuring illuminance at driver s eye point. Figure 7. Illuminance on the road surface. Figure 10. Illuminance at driver s eye point. Figure 8. Experimental comparison of 5 lx lines. is zoned into a rectangular area (60 m 100 m) marked with cross stripes every 2 m. The illumination is measured on each stripe. Figure 8 depicts the experimental results of illuminance on the road surface. In Figure 8, the horizontal axis represents the illumination length (5 lx line), and the vertical axis denotes the beam width. The S-model LED produced the greatest distance of illuminance (100 m), while the HID headlamp provided the widest beam width (28.6 m). The beam widths of the LED headlamps are 21.2 m for the S- model and 21.0 m for the M-model. The M-model LED generates the shortest distance of illuminance (74 m) Glare and Sensitivity Glare Glare To measure illuminance, an illuminometer is placed on the driver s seat at the eye level (approximately 1.2 m above ground) in a subject vehicle, as shown in Figure 9. Then the illuminance is measured every 10 m while the test vehicle approaches closer and closer to the subject vehicle, starting from a distance of 100 m. The measured illuminance levels are presented in Figure 10. Using illuminance at the driver s eye point, adaptation luminance, and the angle between the glare source and the observer s fixation point, the deboer rating (Flannagan et al., 1990; Schmidt-Clausen and Bindels, 1974) is obtained by Equation (1): E b W=5.0 2log C L u po Φ 0.46 C pl, (1) where W : deboer discomfort glare rating E b : Illuminance at eye of the observer (lx) C po : Constant, (lx/min 0.46 ) L u : Adaptation luminance of the observer (cd/m 2 ) C pl : Constant, 0.04 (cd/m 2 ) Φ : Angle between the glare source and the observer s fixation point (min) The deboer rating is graded by nine levels, with a lowest acceptable value of 5. A higher rating means a lower amount of glare. Following Schieber (1998), we employed 1 cd/m 2 for adaptation luminance in (1). The mean value of glare for each headlamp is shown in Figure 11. The results show that all of the tested headlamps yielded acceptable values of the deboer rating. However, because

5 PERFORMANCE EVALUATIONS OF LED HEADLAMPS 741 Table 3. Power spectral distribution and luminance. Main wavelength (nm) Distribution of main wavelength (%) Luminance (cd/m 2 ) Halogen ,210 HID ,470 LED ,071 Figure 11. Results of glare rating. different light sources deliver different colors of light, the evaluation of sensitivity glare must consider both light color and color temperature Sensitivity glare The sensitivity glare is observed while an observer (subject) vehicle is driving, as shown in Figure 12. In this experiment, the sensitivity glare of LED headlamps is found to be higher than that of halogen and HID headlamps because the color coordinate of LED light is more concentrated to the blue zone (see Figure 4). Human eyes become very sensitive to short wavelengths (450 nm to 500 nm) at night (Boyce, 2003; Sivak et al., 2005). Figure 13 and Table 3 summarize glare source spectral power distribution and luminance measurements for LED, halogen, and HID headlamps. We analyzed the spectrum of white LED light sources and found the maximum energy to occur at a primary wavelength of 466 nm. Halogen and HID lamps have main Figure 12. Sensitivity glare test. Figure 14. Results of visibility test. wavelengths of 588 nm and 580 nm, respectively. Similar results also can be found in Bullough et al. (2003). The primary wavelengths of halogen and HID lamps are outside the sensitive band. Thus, it is readily identifiable that light from LED headlamps results in higher glare than from conventional headlamps. This analysis reveals that sensitivity glare is significantly affected by both color and wavelength of light Visibility A visibility test is carried out to measure visual recognition at short and long distance, such as recognition of traffic signs or road studs. The evaluations are described in five levels (level 5: excellent, level 4: good, level 3: average, level 2: poor, level 1: very poor). Figure 14 summarizes the results of visibility tests using halogen and LED headlamps. Visibility at long distance indicates the recognition of objects and road studs placed far from the driver of a test vehicle. The S-model LED shows the highest performance, followed by M-model LED and halogen headlamps. LED headlamps also demonstrated better performance than halogen headlamps at short distances. LED headlamps allow drivers to see more road studs and provide wider beam width. Recognition of traffic signs also increased with the use of LED headlamps. 4. CONCLUSIONS Figure 13. Comparison of typical spectral power distribution for halogen, HID, and LED headlamps. Research on automobile front lighting systems has been ongoing for decades in an effort to reduce pollution and

6 742 B. KANG, B. YONG and K. PARK save energy resources. The development of LED headlamps is one of the many valuable results of this research. We evaluated LED headlamps and compared their performance with that of HID and halogen headlamps. The LED headlamps examined in this study performed well with respect to photometric performance and visibility. The findings herein are summarized as follows: 1) LEDs outperform halogen headlamps and are competitive with HID headlamps with respect to photometric performance, such as luminous intensity and beam width. 2) The color of LED headlamp light does not satisfy the Korean Motor Vehicle Safety Standard. The blue-white color that appears near the cut-off line may cause a discomfort in glare. 3) The glare test and the sensitivity glare show that LED headlamps are still inferior to conventional headlamps with respect to color and wavelength of light. 4) LED headlamps do provide sufficient visibility. ACKNOWLEDGEMENT This investigation was carried out with the support of HMC. REFERENCES Apfelbeck, R. (2007). Advanced optical and thermal solutions for LED headlamp. SAE Paper No Boyce, P. R. (2003). Human Factors in Lighting. 2nd edn. Taylor & Francis. London and New York. Bullough, J. D., Van Derlofske, J., Dee, P., Chen, J. and Akashi, Y. (2003). Impact of headlight glare on peripheral visibility. Progress in Automobile Lighting 5th Int. Symp., Dubosc, C. (2005). Compared future styling trends for lighting systems on major markets. 6th Int. Symp. Automotive Lighting Flannagan, M. J., Sivak, M. and Gellatly, A. W. (1990). Rearward Vision, Driver Confidence, and Discomfort Glare using an Electrochromic Rearview Mirror. UMTRI Transportation Research Institute. University of Michigan. Ann Arbor. Hamm, M. (2005). LED revolution in headlamp optics, mechanics, thermomanagement and styling. 6th Int. Symp. Automotive Lighting, Kang, B., Park, K. and Son, S. (2006). A evaluation on LHPS cooling performance of LED headlamp. Spring Conf. Proc., Korean Society of Automotive Engineers, Kang, B., Park, K., Yong, B. and Kim, H. (2007). Application of LED Headlamps in S. Korea. SAE Paper No Kang, B., Park, K., Yong, B., Lee, H. and Kim, H. (2008). The evaluation of performance on LED headlamps. Fall Conf. Proc., Korean Society of Automotive Engineers, Neumann, R., Dobrus, V., Popelek, J., Sayers, E. M. and Kubena, V. (2007). High efficient LED headlamp design - Styling versus light performance. SAE Paper No Scholl, M. (2005). Global headlamp application with dual filament bulbs. 6th Int. Symp. Automotive Lighting Schieber, F. (1998). Analytic Study of Daytime Running Lights as Potential Sources of Disability and Discomfort Glare under Ambient Illumination Conditions Ranging from Dawn through Dusk. Heimstra Human Factors Laboratories. University of South Dakota. Vermillion. SD Schmidt-Clausen, H. J. and Bindels, J. Th. H. (1974). Assessment of discomfort glare in mirror vehicle lighting. Lighting Research and Technology 6, 2, Sivak, M., Schoettle, B., Minoda, T. and Flannagan, M. (2005). Blue Content of LED Headlamps and Discomfort Glare. UMTRI Transportation Research Institute. University of Michigan. Ann Arbor.