Automatic real-time guidance of laser machining with inline coherent imaging

Abstract

Optical coherence imaging can measure hole depth in real-time (> 20 kHz) during laser drilling without being blinded by intense machining light or incoherent plasma emissions. Rapid measurement of etch rate and stochastic melt relaxation makes these images useful for process development and quality control in a variety of materials including metals, semiconductors and dielectrics. The ability to image through the ablation crater in materials transparent to imaging light allows the guidance of blind hole cutting even with limited a priori knowledge of the sample.Significant improvement in hole depth accuracy with the application of manual feedback from this imaging has been previously demonstrated [1]. However, the large quantity of raw data and computing overhead are obstacles for the application of coherent imaging as a truly automatic feedback mechanism. Additionally, the high performance components of coherent imaging systems designed for its traditional application in biological imaging are costly and may be unnecessary for materials processing. In this work, we present a coherent imaging system design that costs less than a fifth of comparable commercial products. We also demonstrate streamlined image processing suited for automated feedback that increases processing speed by two orders of magnitude.Optical coherence imaging can measure hole depth in real-time (> 20 kHz) during laser drilling without being blinded by intense machining light or incoherent plasma emissions. Rapid measurement of etch rate and stochastic melt relaxation makes these images useful for process development and quality control in a variety of materials including metals, semiconductors and dielectrics. The ability to image through the ablation crater in materials transparent to imaging light allows the guidance of blind hole cutting even with limited a priori knowledge of the sample.Significant improvement in hole depth accuracy with the application of manual feedback from this imaging has been previously demonstrated [1]. However, the large quantity of raw data and computing overhead are obstacles for the application of coherent imaging as a truly automatic feedback mechanism. Additionally, the high performance components of coherent imaging systems designed for its traditional application in biological imaging are costly and...