The problem. Multi-material printers can place different materials anywhere inside a part, but conventional CAD describes only surfaces: each body is one uniform solid. Engineers who want smooth material gradients, graded lattices, or property-driven designs end up writing one-off scripts for a single printer.
The idea. OpenVCAD represents an object as implicit geometry plus named attribute fields, such as color, stiffness, or material fractions, defined at every point. Designs are ordinary Python programs, so they are parametric, reusable, and easy to combine with optimization and simulation. Nothing is voxelized or sliced while you design. When a design is ready, a compiler samples its fields for a specific process, and translation models convert requested properties into what each machine needs.
What it enables. One design can be compiled for several printing processes. Systems and export formats supported right now include:
- Inkjet (PolyJet)Material mixtures
- Filament (FFF)Graded process parameters, mixing printers, and material assignments
- Vat photopolymerizationMulti-material and grayscale exposure
- Direct ink writingMaterial mixing
- Meshes and slicer projectsMeshes and PrusaSlicer projects
OpenVCAD has been downloaded tens of thousands of times, and researchers at more than two dozen institutions use it for medical models, impact-absorbing structures, soft robots, printed electronics, and much more. Its lattice tools have their own page.
My role. I created OpenVCAD and lead its development in the Matter Assembly Computation Lab with Robert MacCurdy. It is the foundation for the rest of the work on my research page.