Error Analysis in Long-Range Test of LED Street Light Characteristics

1 Introduction

In recent years, LED light source has been recognized and valued by its high efficiency, longevity and environmental protection. At the same time, with the rapid development of LED technology in the past 10 years, the related secondary optical design technology of LED has become more and more mature, replacing the existing semiconductor light source. Urban road lighting sources have become a major trend in the field of semiconductor lighting, and have received increasing attention. However, in order to promote LED street lamps in a large amount, it is necessary to have a unified technical standard and measurement standard for LED street lamp products, and stipulate some mandatory performance indicators to ensure the performance and quality of the products. From the current status quo of LED street lamp product technology development, the most important ones are the whole lamp life, the whole lighting effect and the light decay curve.

LED street light as a system, many factors affecting the above performance, such as the characteristics of the power drive, the weather resistance of the optical components, the waterproof and dustproof performance of the whole lamp structure, etc. will seriously affect the light efficiency and life of the system, that is, the LED for light The optical characteristics of the chip or module do not fully describe the performance of the luminaire. Therefore, the aging test of the whole lamp must be performed to obtain data that meets the actual performance of the system.

As the LED street lamp for outdoor lighting products, the reliability must be carried out in the long-term test on site. The time span often takes thousands of hours, and the accuracy of the long-range measurement on the site is affected by environmental and test methods, and it will be compared with the indoor experiment. The test results of the room environment produce large differences.

Figure (1) compares the light fading results of different LED street lights provided by a number of companies organized by the inspection department in the laboratory environment and in outdoor venues.

Figure (1) shows the data of the light decay test of the 22 LED street lamp products for 2000 hours. It can be seen from the figure:
A) The light decay value of the LED street lamp tested in the outdoor field is greater than the light decay value tested in the laboratory environment;
B) If it is based on laboratory tests, most of the light decay is below 5%, and if it is based on outdoor tests, the light decay of these street lamps is obviously too large, and it is not suitable for mass production;
C) The difference in this test for different streetlight products varies from 5% to 15%.

This has caused difficulties for the end user to choose LED street light products, that is, what method should be used to detect, which group of test data should I believe?

At present, in all the standards and norms related to LED street lamps at all levels, strict requirements are imposed on the light decay of LED street lamps. Basically, after 3,000 hours of continuous lighting, the road lighting is not higher than 2%; High light fading tests require that all types of critical factors that cause measurement errors must be analyzed and appropriately compensated.

2 Main factors affecting the accuracy of LED road lighting measurement

For the life expectancy measurement method of LED street lamps, the relevant national standard gives the life expectancy measurement method of LED street lamps. The method stipulates that the LED street lamps are reinforced under the conditions specified by certain technical specifications, and the LED street lamps are monitored at the specified temperature measurement points. The temperature of the outer casing rises. The illumination at the lower point of the LED street light at a certain distance is recorded every 100~300 hours until it reaches 6000 hours. If the LED street lamp life is not reached after 6000 hours, the life of the LED street lamp, that is, the life expectancy, can be extended by the law that the illumination below the lamp is lowered. In the life expectancy measurement and extrapolation calculation, the LED street lamp has an initial value of 1000 hours of illumination, and the time when the out-of-light LED lamp flux drops to 70% of the initial value is the expected life.

It can be seen that the LED street lamp is subjected to light decay and life test, and the time span is more than 3 months. Because the impact of the test site often needs to be tested at the street lamp installation site, as shown in Figure (2).

Therefore, it is inevitable that the test accuracy will be judged and the interference factor will be eliminated. Here we first analyze the main factors affecting the LED road light efficiency measurement accuracy.

1) Power supply stability

In the "Measurement method of integral LED street lamp", the power supply voltage requirement is: during the stable period, the power supply voltage should be stable within ±0.5% of the rated value; when measuring, the power supply voltage should be stable at ±0.2% of the rated value. Within the scope of the fundamental wave frequency deviation shall not be greater than 0.1%, harmonic distortion shall be less than 3%; the power supply voltage of the life test shall be stabilized within ±2%.

In terms of LED driving, constant current driving is usually used, and the fluctuation of current directly affects the luminous flux of LED. In order to verify the performance of actual products, two LEDs are specially designed and have more experience in LED streetlight power supply. The power driver's LED driver has a current control accuracy of ±2% and 3%, respectively.

However, this is only the steady state condition indicated. Here we should pay attention to several parameters: the fluctuation of the test point under the lamp with the steady current changing with the applied voltage, the chronic drift with the running time and the stability with the external temperature.

The laboratory test of a company's 75W LED switching power supply shows that the output current of the switching power supply does not change with the change of the applied voltage, and the stability accuracy reaches ±0.01%, which can be ignored. For the aging test of the 75W high-power switching power supply, after 2400 hours of aging, the power supply output current stability accuracy reaches ±0.02% or more, and the influence of this factor on the steady current accuracy of the switching power supply output can be ignored.

We know that even in southern cities, for a 2000-hour test, the ambient temperature difference is at least 20oC (we discuss the temperature difference effect below we are 20oC), this temperature change will affect the LED switching power supply output current The change.

Figure (3), temperature characteristics of LED switching power supply output current

Figure (3) is the result of detecting the temperature characteristics of an actually used LED street light source. It can be seen from Figure (3) that the relationship between the LED switching power supply and the ambient temperature change is I-1 ·dI·dT=-0.0736%/K, that is, the output current of the switching power supply drops at a rate of 0.7% per 10 degrees when the temperature rises. (The ambient temperature is 25OC as the initial temperature). Since the LED output luminous flux is substantially linear with the input current, when the temperature changes by 20 OC, the current output will drop by 1.47%, and the falling current will directly lead to a decrease in the LED output luminous flux. The LED output luminous flux is linear with the input current. Within this temperature variation, the luminous flux of the luminaire will also decrease by 1.47%.

2) Stability of LED light source

LED light sources are semiconductor devices that are very sensitive to temperature. In the test, it must be considered that the luminous flux of the LED source is affected by the ambient temperature, which is a very important aspect that affects the performance of the whole lamp. The relationship between light emission intensity and temperature is usually obtained from the temperature characteristics of the semiconductor device:

I(T)=I0exp(-T/T1)

Here T1 is the characteristic temperature, and the characteristic temperature of the blue LED device using the GaInN/GaN quantum well active layer
T1 = 1600K.

For a white LED device with a blue chip and a yellow phosphor, the temperature characteristics of the device are deviated from the above theoretical formula because the performance of the device is also affected by materials such as phosphors and fillers. Under normal circumstances, the change of light output of white light with temperature is roughly Φ-1dΦ/dT = -0.2%K-1, which means that the output of luminous flux increases with temperature by 2% per 10 degrees. The rate drops. However, for different chips and different packaging processes and materials, the above parameters will also change.

In the experiment, the relationship between the output luminous flux and the junction temperature of two different LED devices at home and abroad was tested. The results are shown in Figure (4).

Figure (4), the relationship between the light output of different white LEDs and the temperature

For sample A, the relationship between the change of light output of white light with temperature is Φ-1dΦ/dT = -0.22% K-1, and for sample B, the above formula becomes Φ-1dΦ/dT = -0.30 %K-1 That is, the luminous flux drops by 3% for every 10 degrees increase in temperature. Therefore, if the temperature difference exceeds 20 ° C during the measurement, the luminous flux test of the LED luminaire will produce an error of the order of 5% or so.

It can be seen that in the long-time light decay test, the influence of temperature must be taken into account, and the corresponding correction is performed according to the measured temperature. Conversely, if the test is started in the summer and ends in the winter, the attenuation will be greatly reduced, and even an increase in light output will occur.

3) Impact of test equipment

Currently used pavement test equipment, usually illuminance meter or brightness meter, the probe of these instruments is the core to determine the accuracy of optical parameter test. General photosensitive elements (such as silicon photo cells, photoresistors, etc.) have their inherent relative spectral sensitivity. The curve, and the ideal spectral sensitivity curve should be exactly the same as the V(λ) curve of the CIE 1931 human eye, which is the only way to achieve accurate photometric measurements. However, the spectral sensitivity curve of a typical photosensitive device differs greatly from the CIE 1931 V(λ) curve, so a series of color filters need to be added for calibration, thereby generating a spectral response error f1.

According to CIE 69-1987 and China JJG245-2005 illuminance measurement and verification procedures, the classification technical requirements of different grades of photometric probes/photometers are shown in Table 1.

For the test equipment of general enterprises, most of them are first-class precision, and their V(λ) mismatch error f1 is 6%. Therefore, LED street lights with different light colors will inevitably cause certain measurement errors due to mismatch. Since f1 describes the maximum error of the probe, it is not the same value in all wavelength ranges. The total error of the last photometric measurement should generally be less than this value.

Set at a certain wavelength, the V(λ) mismatch error of the probe is f1(λ), and the total error of the photometric measurement is:


φ (λ ) is the light intensity of the measured light source at the wavelength λ. Since the emission spectrum of the white LED is mainly concentrated in the two main bands of yellow-green and blue, the yellow light has the greatest influence on the luminous flux, so the probe is at this wavelength. Range mismatch errors are especially important for the accuracy of LED photometric measurements.

In addition, the influence of the temperature coefficient of the photodetector is the most important. According to the Chinese JJG245-2005 illuminance measurement and verification procedure, even if the standard-grade probe is used, the temperature coefficient is 0.2%, which means that the time span is For the 2000 hour test period, this one will also cause a 4% error in the luminous flux test of the LED luminaire under the condition that the ambient temperature changes to 20 °C.

In order to systematically study the temperature characteristics of the illuminometer, we conducted a temperature characteristic test on the optical probe and the whole illuminance meter of a company's first-class illuminance meter. The results are respectively E1 and E2 curves in the following figure (5). .

Starting at 20oC, the E1 curve in Figure (5) shows the temperature characteristics of the photometric head, which includes V(λ) correction filter, cosine correction glass, meter resistance and silicon photocell, which rises with temperature. The output signal has a falling rate of 0.062% K-1. The E2 curve shows the temperature characteristics of the overall illuminometer including the probe and the A/D conversion circuit and the display circuit behind the illuminometer [[5]. The overall illuminance decreases the illuminance rate by 0.374% K-1 with temperature. It can be seen that the temperature change of the rear A/D conversion and display portion has a greater influence on the illuminance value. Tested with this illuminance meter, the measurement error caused by the temperature difference of 20oC is 7.48%.

4) Impact of test methods

In the field measurement, it will inevitably be affected by various types of stray light. The main stray light includes:
1. The illumination of vehicles on the road;
2. Optical radiation from adjacent street lamps;
3. Interference from other light sources on the roadside;
4. The interference of the street light source on other objects;

The above-mentioned stray light is essentially unsteady. For example, the light radiation of adjacent street lamps varies with the position of the street lamp and the distribution of the road light type; when the roadside green plants, buildings, billboards, etc. are blocked or When the reflected light is radiated, the interference of such light is also extremely unstable.

To eliminate the effects of such unstable stray light, an auxiliary isolation can be used in the field test, as shown in Figure 6.

Figure (6), Schematic diagram of the isolation cylinder used to eliminate the effects of stray light

The inside of the isolation cylinder is coated with an absorbing light coating, which can effectively isolate the external interference light, thereby ensuring effective acceptance of the light radiation of the measured street lamp directly above. In the figure, d is the surface diameter of the illuminometer, H is the height of the street lamp, and D is the aperture of the isolation tube. From the perspective of isolation of external interference, the size of D should be as close as possible to the size of d, because LED street lamps are mostly An array of LED modules, the scale of the equivalent luminous surface can not be ignored, the structure of the isolation cylinder needs to ensure that the light of all the LED modules on the luminous surface of the street lamp can be effectively accepted, namely:


5) Impact of environmental pollution

Outdoor LED lamps will inevitably be contaminated by dust accumulation, colloidal suspended matter and other impurities. This pollution will directly lead to the decline of the light efficiency of the lamps when the light-transmitting surface of the lamps appears, and the degree of this effect will also be installed with the street lamps. The location, the time of use on the road surface, and the handling of the surface of the streetlight vary.

In order to test the influence of street surface pollution on the photometric test, we tested the fixed-point illumination with time on the LED street light on the normal road. Figure 7(a) shows that the 10 street lamps on the same road are not cleaned on the surface. In the case of the test results, in Figure 7 (b) shows the results of the last measurement after the illuminating surface of the luminaire. There is a clear gap between the two. As can be seen from Table 2, after 4,500 hours of road surface operation, the illumination test error caused by surface contamination of the street lamp averages 7%.

The above results are obtained for the test of LED street lamps with a flat surface structure. For some LED street lamps with a light-emitting surface and a secondary optical lens surface and having a concave-convex structure, the effect will be more serious and will further affect the light distribution. . Especially in the case of highways and tunnels with large traffic volume, the surface of the lamps is often seriously polluted by oil. It is difficult to effectively clean the light-emitting surfaces of these LED street lamps with concave and convex structures by the conventional cleaning methods for tunnel lamps. .

3 Summary

From the above experimental analysis, it can be known that the long-range measurement of the whole lighting parameters of LED street lamps will be affected by many external factors, and the influence of temperature can not be ignored. When the temperature rises by 1oC, the measurement errors introduced by the driving power source, LED light source and illuminometer are -0.0735%, -0.25%, -0.374%, respectively. When the temperature rises by 20oC, the total measurement error caused by the above errors Will reach -13.95%. The illuminance test error caused by surface pollution of street lamps is also quite serious. Under the application environment of urban main roads, the test illuminance drop caused by surface pollution after 4500 hours of road surface operation is also about -7%. Looking at Figure (1), it is not difficult to explain the difference between the indoor and outdoor test data of different manufacturers by 5%-15%. Of course, our data is only to study a certain kind of power source, light source and illuminometer, different power sources, different LED light sources, different illuminometers are affected by temperature.

From the above research conclusions, it is necessary to accurately obtain the photoelectric parameters of the LED streetlights in the long-range measurement with large time span, and it is necessary to eliminate the influence factors of various non-LED light sources as much as possible, especially in outdoor field tests. The impact of various externalities is also more serious. By cleaning the luminaire and standardizing the test method, it is possible to reduce or eliminate the error caused by environmental pollution and test methods. However, the switching power supply, LED light source and measuring illuminometer in the LED street light system are greatly affected by temperature. If the test light attenuation is less than 3%, the measurement accuracy should be better than 1%, so standard test instruments should be used; the temperature influence must be controlled (the change of ambient temperature within the test time span ΔT<3°C) Or give necessary compensation; the collection of light must avoid the interference of various stray light; at the same time, the serious influence of the surface pollution of the lamp on the test must be considered.

references:

1. E.Fred Schubert, Light-Emitting Diodes, second edition[M], Cambridge university
Press,2006:191
2. A. Zukauskas, Introduction to Solid-State Lighting [M]. New York, Wiley, 2002.
3. Datasheet Lumileds Luxeon emitters, Lumileds website [online]: http:// ww.lumileds.com
4. Pan Jiangen, Yan Fangsheng. High-precision photometric probe for measuring LED and its performance evaluation [J]. "LCD and Display",
Vol. 20. NO. 2, Apr. , 2005
5. Hou Wenhui. Design of high-precision illuminance meter [D]. Dalian: Dalian Institute of Technology. 2007: 3-8.
6. Xu Yan.Characteristics of illuminance meter and quantitative evaluation of measurement error[J].《现代测测试"》vol3.2002 

The above content is provided by WOSEN. WOSEN is a professional manufacturer and supplier of Led Flood Light, Led Street Light, Led Solar Light, etc. For more information, please visit https://www.wosenled.com/ or contact admin@wosenled.com or WhatsApp +86-13425434349