During the measurement of luminous flux with integrating sphere, comparing with common lighting source, the luminous flux test accuracy of LED lighting source have given huge challenge to the testing equipments. On the one hand, LED light source have outstanding directively characterizes than other ordinary light sources. Normally it wouldn’t uniformly emit lighting at all the direction on the whole space. This feature makes LED’s direct light distributed unevenly on the surface of integrating sphere, which directly causes different direct reflective light of LED for different reflective characteristics of the detector. Because the position of detector port and baffle is fixed, different reflectance distribution directly presents as signal fluctuations. In the conventional measurement system, the actual measurement value also shows great differences for LED of different forward divergence angle, the same LED of different orientations, different positions in the same direction and others, even if the nominal luminous flux is the same. According to the results verified by customer, in the conventional LED measurement system, the effect of LED placement direction on the results of luminous flux measurements is often more than 50% (difference between the maximum and minimum signal of the same LED measured in different directions).
When measuring different LED of different lighting angle, the different distribution on the surface of integrating sphere causes the effect of the direct reflectance distribution on the detector is different, thus directly affect the difference of accuracy between the two measurements. As shown in the picture:
On the other hand, LED measurement system usually uses a halogen lamp as standard light source. The standard light source is totally different from LED lights on the appearance, luminous distribution or spectral characteristics side. Therefore, the difference must be corrected by the self-absorption coefficient.
One of the important reason of LED’s directionalities influence the testing accuracy focus on inner surface reflectance characteristics of the integrating sphere. In the common LED measurement system, both the reflectance and lamberts characteristic of integrating sphere surface coating are not very satisfactory. One is low reflectivity, the other is bad diffuse reflection characteristics. One result of low reflectivity on the surface of integrating sphere is the direct lighting of LED gradually decays after a few times reflection. In the whole process of light mixing, the direct lighting and direct reflected light accounts for a large proportion and plays a dominant role. However, at some conditions, the low reflectance material will produce a strong shadow effect on the detector at the back of baffle. What results in inaccurate measurement is nothing else but the reflected light and shadow effect.
Moreover, lower diffuse reflectance will seriously cause the signal attenuation. During the measurement, since the continuous reflection of light inside the integrating sphere and the reflection decays each time, the effects of high or low reflectance on luminous intensity can be strengthened after multiple reflections. Reflecting the light in the ball 15 times, for example, if both reflectance difference is 5%, then the attenuation of the signal may be double or more. However, the difference of reflectance inside the integrating sphere is much more than that.
Currently, LED testing system has not been used LED standard light as a standard light source but a calibrated halogen with stable driver during the measurement. Due to the appearance of standard light source and the tested LED varies widely, LED fixture’s absorption effect to light, and the difference between mounting position of standard light source and the tested LED, which are all the important factors affecting the measurement accuracy of the results.
Lisun developed a IS-*MA new design integrating sphere with testing hold base. Compared to the “massive assembly” production technology of traditional integrating sphere, IS-*MA adopted a molding technology to produce the integrating sphere which shape fully complies with 4π or 2π spherical structure, and used coating of high reflectance and diffuse reflectance, and designed the luminaries opening position right towards the detector. With this improvement, even it is under extreme conditions which using LED with strongly high directionality or placed LED with extreme conditions, the measurement results are still maintained a good consistency.
LPCE-2 system uses a calibrated halogen lamp as standard light source. Meantime auxiliary lamp is available for alternative solution, which use to compensate for the impact of the difference of clamp of measured LED and standard lamp on the measurement results. LPCE-2 system specifically tests for LED measurement accuracy problem as described above. Test conditions are as follows: adopting high brightness green 5LED, which power is about 0.35W, emitting angle is about 30°.
LPCE-2 system uses nine measuring position, representing the possible LED placement position shown as Figure III.
Figure III Different Placement Position of LED
The relationship between the lumen and the LED placement are as Figure IV. From the test results, even it is in the most extreme case which LED towards or back towards the opening of detector, the luminous flux measurement result is still less than 5%, which is a very good test results. In the practical application, since the LED would not be placed as such extremely situation, it generally uses a simple test fixture in testing. Under the case of including positioning error, measurement results error of luminous flux in the same position is less than 0.1%. However, the test error of LED luminous flux measurement repeatability is much less than 0.1% during the actual testing.
Figure IV Lumen Value for different measurement position of LEDTags：IS-*MA , LPCE-2(LMS-9000) , LPCE-3