Journal of Materials Processing Technology 219 (2015) 17–27 Contents lists available at ScienceDirect Journal of Materials Processing Technology jo ur nal ho me page: www.elsevier.com/locate/jmatprotec An analytical model for scanning-projection based stereolithography Mohammad Mahdi Emami a , Farshad Barazandeh a,∗ , Farrokh Yaghmaie b a Department of Mechanical Engineering, Amirkabir University of Technology (Tehran Polytechnic), 424 Hafez Ave., Tehran 15875-4413, Iran b New Technologies Research Center (NTRC), Amirkabir University of Technology, Valiasr Ave., Tehran 15916-33311, Iran a r t i c l e i n f o Article history: Received 7 August 2014 Received in revised form 2 November 2014 Accepted 2 December 2014 Available online 9 December 2014 Keywords: Additive manufacturing Digital micro-mirror DMD Dynamic mask Large area exposure Maskless stereolithography Pixel based modelling a b s t r a c t There is a growing need for effective small scale production methods. Projection stereolithography (PSL) is a technological response to such a demand. In PSL, Experience shows a decrease in resolution as area of exposure increases. A relatively novel hybrid method, scanning-projection stereolithography (SPSL) is presented in this work. This method is based on previous work by a number of authors, utilizing a combination of scanning and projection to manufacture large parts with relatively high-resolution. A modelling method to investigate the total energy received by individual pixels on resin surface is considered for both PSL and SPSL. The modelling shows near identical energy distribution for both methods. The modelling results were attempted to verify experimentally. Four patterns with circular and rectangular features were exposed with both methods. The resulting cured layers were compared via microscopic observation and measurements. Sample measurements show SPSL has a slightly better resolution using an inherently non-uniform exposure system. In large area exposure, SPSL provided less stitching and overlap issue. © 2014 Elsevier B.V. All rights reserved. 1. Introduction Now a day’s additive manufacturing like rapid prototyping (RP) for small scale production and modelling studies are becoming a part of production. These methods are capable of fabricating com- plex shapes. Scanning-based SL (SSL) and projection-based SL (PSL) are two different methods distinguished by their patterning pro- cess. A good review of SL systems is presented in several books including those published by Gibson et al. (2010). There are dis- tinct advantages and disadvantages associated with each one in terms of resolution, layer thickness, accuracy, cost efficiency, and throughput. In SSL a focused laser beam with small diameter (typi- cally 100 m) is scanned to fabricate the desired pattern. For large areas, the laser spot is moved by fluctuations of a Galvanometer- mirrors (Jacobs, 1992). PSL uses a spatial light modulator (SLM) as a dynamic mask for generating a 2D pattern with micro-scale resolution. Digital micromirror device (DMD) is a common SLM used as dynamic mask generators. DMD is a device consists of M × N array of individually addressable -mirrors. Each with a size selected from 17, 13.7 and 10.8 m 2 (Hornbeck, 1997). Each individual -mirror can rapidly rotate to provide ON/OFF light switching. ∗ Corresponding author. Tel.: +98 21 6454 3442. E-mail addresses: mehdi.emami@gmail.com (M.M. Emami), fbarazandeh@aut.ac.ir (F. Barazandeh). PSL enables creation of features smaller than 10 m and is con- verging to projection micro stereolithography. There are a number of researchers focused on improving the vertical and horizontal resolution. Sun et al. (2005) introduced a solidification model to predict the fabrication results of line patterns and investigated cross-talk as well. Limaye and Rosen (2006) used the compensation zone approach to avoid print-through errors. Also, Limaye and Rosen (2007) and Jariwala et al. (2009) reported a model to predict solid- ified layer thicknesses based on experimental observation in order to optimize the vertical resolution. Zhou et al. (2009) introduced pixel blending strategy to intelligently control pixels’ grayscale level to achieve much higher horizontal resolution. Quality of fab- ricated parts by Zhou showed an improvement especially for small features. Recently, Kang et al. (2012) has developed a pixel-based solidification model for PSL to optimize the horizontal resolution by predicting the intensity distribution. Compared to SSL, PSL offers various advantages in the fabrica- tion of 3D freeform structures with high resolution at high-speed. A key challenge using PSL is the limited resolution and small array size of DMDs. Projecting a pattern with a DMD array over a larger surface will result in a larger pixel size and as a result a lower res- olution. For typical XGA DMDs (1024 × 760), the platform size for micro (10 m × 10 m) and macro (200 m × 200 m) resolution are 10.24 mm × 7.28 mm and 204.8 mm × 152 mm respectively. In other words, PSL is impractical when the whole pattern does not fit into the projection area for a specific resolution. http://dx.doi.org/10.1016/j.jmatprotec.2014.12.001 0924-0136/© 2014 Elsevier B.V. All rights reserved.