In-line polariscopic checking of plastic molded-injected lenses: preliminary results

Plastic injection molded lenses have improved its performance and, nowadays, are as usual as glass lenses in image forming devices. However, the manufacturing process induces the surface generation and the material transformation in the same stage. Moreover, the process also includes an annealing stage to remove the internal stress with temperature cycles but only works up to a certain level and not beyond, leaving relevant traces for high values. During the manufacturing process of a plastic lens, a liquid-solid phase transformation occurs, and in this transition not all the volume of the lens achieves the same density. This change of density is translated into a local change of refractive index that can be expressed as a retardation phase plane using the Jones Matrix notation. The detection and measurement of the value of the retardation of the phase plane is thus the clue to manufacture good and controlled quality plastic lenses. We have tested an in-line polariscopic arrangement to obtain a 2D map of the tension distribution in the bulk of the lens. This test is performed in the first 30 seconds after the injection molding process for two main reasons: first the stress values are still high because the lenses do not have enough time to relax the internal tensions and obtain the final shape, and second, we can remove the wrong lenses in the first moments and introduce only the good lenses in the annealing stage. The proposed instrument is based in a transmission polariscopic arrangement. A collimated light beam is used to illuminate the sample, once the light crosses the sample, it is collected with an afocal system and the image is recorded in a CMOS sensor. Selecting an afocal system to capture the image is a useful decision because the lateral magnification can be maintained when small changes in the sample position are introduced. However the produced lenses can vary their focal lengths from on series to another. To avoid problems with the change of the focal length, the lens is introduced in a matching index and the polariscopic measurement is done. The proposed polariscopic arrangement uses two lineal polarizers, one acting as polarizer and the other acting as analyzer. This system instead of using one lineal polarizer and a lineal polarizer with an extra lambda/4 plate provides us an extra degree of freedom, enabling the possibility to put a certain degree of polarization in a well determined position of the lens, in our case the center of this lens. The aim of this study is to select the minimum number of sets polarizer-analyzer and the right wavelengths to obtain a sure selection of the right lens. The preliminary results show that use two different wavelengths 470 & 627 nm is a good option to obtain a robust image. The second free variables that must be adjusted to obtain good values is the minimum number of set polarizer-analyzer necessary to obtain confident results. In our first tests it seems that recording only at 0, 15, 30 and 45 degrees is enough to get good results. Mathematica description and first results for a PMMA lens are presented, however the number of measurements must be diminished to obtain an easy in-line implementation