In-depth analysis of LED fresh light mixing solutions and simulation technology

In order to meet the lighting needs of fresh foods of various colors , the spectrum of fresh food lamps has become diversified. Although LED light sources with special spectra are theoretically feasible, it is actually impossible to have LED products with every special spectrum ; therefore, many fresh food lamps still use mixed light solutions. However, in order to obtain a suitable mixed light solution, the usual calculation method is very cumbersome; in this article, the existing white light LEDs and color LEDs are reasonably selected, and the mixed color spectrum is obtained by simulation using LightTools. Through the analysis and evaluation of color parameters, the appropriate mixed light solution and effect indicators of fresh food lamps are obtained.

　　With the further development of solid-state lighting, people's requirements for lighting quality are getting higher and higher. Many lighting fields have put forward personalized spectrum requirements. Fresh light lighting in food is a good example. We know that under low CRI and high color temperature lamps, human faces will appear pale and lifeless. This is also the case with food. If inappropriate lighting is used, fruits and food in supermarkets will become weird in color and unappealing, which will not only affect business, but also cause waste because the goods are out of shelf life. Under the appropriate fresh light, the color of food can be better restored, making it look fresher and more delicious, making customers more willing to buy, so fresh light is very important in the field of food lighting. LEDs have become ideal light sources in applications because they can easily obtain various spectral colors.

　　The existence of multiple monochromatic chips and blue chips with special phosphors can realize most spectrums. However, for a single LED, not every spectrum of LED will become a product, because the market demand for each special spectrum of LED is generally not very large; even if there are products, the price will be relatively high, and there may be supply problems. Relatively speaking, the mixed color solution is more feasible, because only conventional LEDs need to be selected, so the light source cost is lower, the solution is flexible, and the supply is more reliable.

　　We usually use color coordinates, CCT, CRI and CQS to evaluate color. Among them, CRI has color rendering indexes of 7 special color samples in addition to the conventional 8 standard color samples. These 15 color samples are relatively unsaturated color samples. The average value of the first 8 parameters is the color rendering Ra that we often use. For example, R9, which represents red light ("strong red"), is not included in the representation of Ra. Therefore, even if the LED has a large Ra, as long as its spectrum does not have enough red light, the value of R9 will be very low. NIST (National Institute of Standards and Technology) also found that even if a light has good color rendering for unsaturated colors, it may have poor color rendering for saturated colors; NIST found that as long as some saturated colors are selected as a new set of color samples, the accurate representation of color rendering can be guaranteed, and a new method for representing color rendering, CQS (Color Quality Scale), is proposed. It uses 15 saturated colors distributed in the entire visible spectrum as its color samples. This article will use these two sets of methods to represent color rendering to analyze the results after color mixing.

　　Given a spectrum, we can get the above-mentioned color information, so the spectrum covers more comprehensive color information. For mixed color lighting solutions, the spectrum information is very important, but how can we obtain it more conveniently?

　　“First, although it can be obtained through measurement, it is necessary to make a sample lamp, which includes steps such as customizing the PCB, preparing the heat sink and purchasing the power driver, welding and assembly, and finally measuring. The process is complicated and time-consuming, and it is difficult to adjust the plan again. The method is not flexible.

　　“Secondly, the calculation method is also very cumbersome. Spectral data is usually normalized, and the ordinate is related to mW (not lm). In addition, the radiation characteristics of LEDs are also different. For example, white light LEDs are usually characterized by luminous flux (lm) and red light by radiant power (mW), making the calculation more complicated.

　　“Finally, the simulation method in this article, with the help of LightTools simulation, can not only easily obtain the mixed color spectrum results, but also obtain other color parameters to help us analyze and evaluate the solution.

　　Of course, the optimal spectrum required will vary depending on the object being illuminated. For example, when lighting a deep red apple, the fruit vendor hopes to get the right amount of deep red light mixed in to make the apple appear rosy and delicious. At the same time, it should be emphasized that there are many types of red light. When choosing a red light source, you must first ensure that it has the correct main wavelength and spectrum information so that the mixed light solution matches the lighting requirements.

　　In order to achieve the desired color mixing spectrum, we must first select the appropriate LED light source. OSRAM has a wide variety of color light products (including the high-power OSLON series and the medium-power DurisP5), especially in terms of red light, there are three bands of products; white light LEDs also have three color rendering indexes and various color temperatures to choose from. So this article takes its products as an example, selects red light LEDs with appropriate main wavelengths and white light LEDs with appropriate spectrums, obtains their spectral information, and simulates them through LightTools software.

We selected the red light (HyperRed) LED (LH) 　　with the largest dominant wavelength (640nm) and the white light LED (LCW) with a color rendering index of 80, and used the (mLCW+nLH) color mixing solution.

　　1. Obtain the spectrum data of the LED. Usually, the specification sheet has a normalized spectrum graph, and then the spectrum data is obtained; if there is no spectrum graph, it can be easily obtained by clamping the LED sample with an existing fixture and using an integrating sphere measurement system.

　　2. Input the spectrum data of each LED in LightTools. Simply copy the spectrum data obtained in the Excel spreadsheet into the software. And you can check the spectrum input results through "SpectralRegionChart", as shown in Figure 1.

　　3. Enter the radiant power/luminous flux of each LED in LightTools. In the datasheet, white LEDs are characterized by luminous flux (lm), while red LEDs are characterized by radiant power (mW). Here we do not need to convert the units. Simply select luminous flux ("PhotometricFlux") or radiant power (RediometricPower) in "Emittance" in the software and enter the values ​​respectively. Usually when entering brightness information, the effects of drive current and junction temperature need to be considered, but here as an example, typical current drive is used and the effects of factors such as junction temperature are ignored to simplify the process and highlight the key points.

　　4. Run the simulation and analyze and evaluate the spectral results. Of course, you may not get the most suitable color mixing solution the first time, so you may need to go back to step 3 and readjust the power/luminous flux ratio, or even add or change the type of LED.

　　5. Select the appropriate color mixing scheme and the corresponding color and brightness information.

　　Figure 1. Input and spectral view of white light and deep red LED light source spectrum data

　　Figure 2. CRI and CQS results for white light LEDs

　　By analyzing the spectrum of white LED light alone, we can find that although its Ra=83, its R9 value representing the red light component is only 11, which means that the light source has a poor ability to restore the color of this red object and is not suitable for lighting red objects. This is why the fresh light for red food needs to be rich in a certain amount of red light, which also highlights the importance of fresh light. Of course, red light cannot be added excessively, otherwise, although the red restoration ability is very high, it will greatly reduce the color rendering of other colors.

　　Assuming the ratio of white LED to red LED is k=M:N, the number of white LEDs should be set to k times the white light flux. We assume that each red LED has a radiant power of 0.3W and each white LED has a radiant power of 100lm. If 500lm of white light is input, then k=5.

　　Figure 3. Mixed light spectrum and its CRI and CQS results

　　Add "Farfieldreceiver" and run the simulation. The results show that this solution adds the red light spectrum well, CCT=2750K, and obtains better CRI (92) and CQS (90). The indexes that characterize the red reduction ability are R9=70 and VS1=90. The main reason for the difference between the two is the different selection of red standard color samples. The results have higher values ​​than white light LEDs, indicating that this fresh light solution is more conducive to the color reduction of red food; at the same time, the color rendering index of other color samples also retains higher or higher values. While achieving the red reduction ability of the fresh light, it also improves the overall lighting quality.

　　in conclusion:

　　The requirements for lamp spectra in fields such as food lighting are becoming more and more diverse, but it is difficult to mass-produce light sources with customized spectra, and they are expensive, so a mixed color solution is usually used. Due to the existence of the visibility curve and the light source spectrum curve, the calculation between the radiant power and the luminous flux is very troublesome. This article uses the existing tool LightTools to simulate the mixed light spectrum through simple settings. In addition to obtaining the spectrum, a lot of other color information can also be obtained; the results show that the lighting effect obtained by this solution is very good, with a color rendering index of more than 90, rich in deep red, and suitable for many red food lighting.