The Main Influencing Factors Of LED Measurement Uncertainty And Corresponding Correction Method

As a new generation of light source, LEDs have many different characteristics compared with traditional light sources and radiation sources.The measurement, applications and research about LEDs can not rely on the traditional means, so we must explore new paths for it.CIE-127 report pointed out:In the measurement of LED, if the responsivity across the entrance aperture of the photometer is not uniform, it can cause significant errors, and in the measurement of different colors of LEDs, even if the detector meet to f1'<3.0%, we also need to consider such photometers should be furnished with individually measured relative spectral responsivity data so a spectral mismatch correction should be applied for the measurement of color LED. Furthermore, the LED's total luminous flux is very small, the measurements are made on LEDs at relatively short distances, and the LEDs can't be treated as a point source, so the measurement conditions must be strictly controlled. CIE-127 report provides some advices on the measurement of LEDs, but no intensive experiment study on the uniformity of detector's responsivity and the spectral mismatch correction. This paper has considered the main influencing factors in the measurement of LEDs and focus on the above two influencing factors.First, a measurement device for accurately measure the distribution of detector's responsivity was developed. It includs the mechanical design, circuit design, control programming, data analysis and processing and corresponding correction method for exploration. The device can also be used to measure the distribution of small-scale illuminance field after a simple transformation. Experiments show that the responsivity of photometer with diffuser is not uniform, and some LEDs have a narrow beam angle or irregular intensity distributions. When using this kind of photometer to measure a white LED, the measurement error is over 2.0%. An error correction model after rigorous mathematical calculations to correct the measuring result has been developed. This correction method improved the LEDs measurement method.Secondly, a miniature integrating sphere was designed to improve the uniformity of detector's responsivity. Experiments show that this method reduces the measurement error root in the uniformity of responsivity distribution. But the using of integrating sphere make the signal to noise ratio significantly reduce and many other problems.Finally, Spectral distribution of LED is quasi-monochromatic, which is different from the spectral distribution of standard lamp. Due to the fact that some LED spectra are peaking at the wings of the V(λ) function where the deviation makes little effects on spectral mismatch correction but can cause large errors in the case of measurement of single-colour LEDs. I use some excellent photometers to measure the different colors of standard LED. Experimental results show that the spectrum mismatch error of every photometer in LED measurement can not be ignored. But we can reduce the spectral mismatch error by choosing a same color of standard LED to calibration photometer before measuring the LED. These experiments illustrate the importance of spectrum mismatch correction in measurement of LEDs.This paper analyzes the main influencing factors in the measurement of LEDs, and focus on uniformity of responsivity across the entrance aperture of the photometer and the spectral mismatch correction. It presents some reasonable suggestions and methods in measurement of LEDs. It provides the technical basis for LED testing regulations, and it is a supplement of CIE-127 and improves the measurement specifications of LEDs to be more perfect.