Laser Scanning Modalities Revisited

The aim of this study is to compare advantages and disadvantages of some of the most common laser scanning modalities: raster scanning (that is considered the gold standard in laser scanning systems), spiral, Lissajous, and, Rhodonea curves-based scanning (the latter generated for example by Risley prisms). Aspects such as linearity, field-of-view (FOV), and fill factor (FF) have been explored for these modalities. The first two have the potential of linearity and distortion-free, with the drawback of being slow. The latter two are inherently non-linear but with high scan speed. In this study, simulations were performed with MathCAD 15 for spiral, Rhodonea curve-based, and Lissajous scan patterns. Raster and Lissajous scan patterns were experimentally validated with in-house two-dimensional (2D) galvanometer scanners (GS). For spiral scan patterns, Fermat, Archimedean and constant linear velocity (CLV) scanning were examined, in terms of uniformity and distortion. The first one has distortions at margins, with the number of generated spires equal to the frequency; the latter two are uniform and presenting distortions in the center of the scanning pattern, due to increased beam density in that region. The Cornu spiral was also briefly approached. The main focus has been on Lissajous based scan patterns; its FF analysis was carried out for the <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$f_{x}=32$</tex> Hz frequency on the horizontal axis and <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$f_{y}=1$</tex> to 32 Hz adjustable frequencies on the vertical axis. The cases of <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$f_{y}=30$</tex> and <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$f_{y}=31$</tex> Hz were considered, to compare the even- and the odd-numbered case. For the latter, the FF was double. The frequency ratios <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$f_{y}:f_{x}=3: 2; 7: 3$</tex> and 7: 10 were considered for validations, with a good match between simulations and experimental results.