Volumetric Additive Manufacturing (VAM) is a recently-conceived method of printing into a fixed volume of photosensitive resin in a single lithographic step. It has multiple advantages over prior printing methods. Printing into highly viscous or solid materials is possible, enabling a new range of material properties. Print times are dramatically reduced, non-contiguous parts become possible, and there are no inherent layering effects that plague other methods via inhomogeneity, anisotropy, and cosmetic defect of final parts. However, there remain significant challenges that limit the potential of VAM. 1. Poor dose contrast of tomographic reconstructions leaves no room for errors in timing, in optical uniformity, etc., without impacting print shape fidelity. 2. VAM suffers from large striations, similar in appearance to layering, and with similar penalties. Here, we present a VAM image computation algorithm that significantly improves the contrast of applied dose, along with other performance metrics. This method takes a fundamentally different approach than prior algorithms by algebraically optimizing a model of the printed object instead of directly optimizing projection images. We also present a simple and effective method of eliminating striations in VAM via the addition of a latent cure step. We show that this method, facilitated by the high contrast reconstructions from our image computation algorithm, preserves theoretically perfect print shape fidelity. We conclude by discussing extensions and future work.
|