Dec 19, 2018 | By Cameron

A team of researchers out of Columbia University’s Department of Engineering have duplicated the grains of a real wood plank in a 3D printed plank using destructive tomographic imaging, stochastic dithering, and voxel printing. Don’t worry, all that science talk make sense soon. Multi-material 3D printing isn’t particularly new; if a designer wants different components of a single 3D print to be different materials, that’s easily done. A gear can be stiff, a flap can be flexible, and the housing can be transparent to reveal a gizmo’s inner workings, and it all takes just a few mouse clicks for a designer to instruct the 3D printer to produce each component in its specific material.

Voxel printing is 3D printing based on the smallest possible feature that a 3D printer can handle, in this case that’s a single droplet of photoacrylate resin jetted from a Stratasys J750 3D printer. These types of PolyJet printers are capable of 3D printing up to six different materials at a time, and they function much like standard inkjet paper printers, so their resolution is very high. Each droplet of resin is a voxel and each voxel can be a different material, so gradients of colors and textures can be 3D printed. Generally, the outside of an opaque object is the only part of a 3D print that’s colored because it’s all that’s ever seen, and it’s simple for a designer to apply color to the outside of a digital model and then 3D print it.

But how does a designer replicate the interior of a material with an anistropic internal structure? Anistropic materials are those like wood, which have different physical properties when measured from different directions. The first step is to determine what the interior actually looks like, and for this study the researchers (Fabian Stute, Joni Mici, Lewis Chamberlain, and Hod Lipson) employed destructive tomographic imaging. Essentially, they shaved thin layers off of a plank of olive wood, taking pictures of each layer; stacking the 230 images on top of each other approximated what the internal grain of the plank looked like.

Next, they had to convert those images to a bitmap-based format that’s more voxel friendly, and that’s where the stochastic dithering comes in. With 2D pixels, dithering is the process of simulating a color (orange) by displaying two different colors (yellow and red) on touching pixels. This is necessary to accurately depict smooth color gradients and motion on screens made up of pixels that can produce only one color at a time. Stochastic dithering is applying that same concept to the stacked images of the wood, creating a 3D gradient between each image layer to form a cohesive voxel-based 3D plank.

The research team wrote a script to convert the RGB (red, blue, green) stacked images into the CMYK (cyan, magenta, yellow, black) color space used by the J750 3D printer, and the final step was printing them. Two planks were 3D printed in different shades and with slightly different print settings; both were then cut or smashed open to reveal their grain. An alligator was also 3D printed with the wood grain to demonstrate applying the internal structure of one object onto an arbitrary geometric shape.

Olive wood was chosen because it exhibits anisotropies that closely map to its visual properties, meaning its physical properties can be predicted by the way it looks. The direction of the grain is visually apparent, and strength goes against the direction of the grain, therefore strength can be inferred by the look of the grain. Such properties provide designers with a starting place for researching and developing software and workflows for multi-material anistropic 3D printing. While the Stratasys J750 is technically capable of mixing materials with differing durometers (hardness), the team focused just on different colors to prove the concept.

Their effort reflects the large gap between the technical limits of advanced 3D printers and the design software used to generate 3D models for those printers. The J750 has a build volume that contains 760 billion voxels, but the trick is commanding those individual voxels in a seamless, intuitive way that isn’t labor or computationally intensive. If you know how to do that, Stratasys would love to hear from you.



Posted in 3D Printing Application



Maybe you also like:


Pipo wrote at 12/20/2018 5:31:50 AM:

Since is the best!!!

Leave a comment:

Your Name:


Subscribe us to Feeds twitter facebook   

About provides the latest news about 3D printing technology and 3D printers. We are now seven years old and have around 1.5 million unique visitors per month.

News Archive