Graded Formations

Graded Formations is an additive manufacturing process that utilizes the benefits of printing with liquid materials and a novel digital design and fabrication workflow to create objects with functional stiffness gradients, inspired by the logic of natural systems.

This research was undertaken as my Master's thesis project, coordinated by the Institute for Computational Design and the Max-Planck Institute for Intelligent Systems in Stuttgart, Germany.

To model material property gradients, the approach was to use a bitmap image with the color value converted to a 'material ratio' that designated the amount of each material to be mixed during a fabrication step. A computational tool was developed to automatically calculate fabrication data based on an input gradient design.

Fabrication was executed with a simple, scalable hardware setup consisting of a material delivery system, a positioning system, and control system.

The software workflow was built using Grasshopper. This pipeline offers a graphical input of color gradients to assist this material attribution, as well as methods that allow the inclusion of fabrication parameters during the design process, allowing the designer to intuitively combine different material properties and take advantage of the fluidic characteristics of the material.

Fabrication path and parameters such as deposition rate could be tuned to add another means of geometric variation. This process enabled a disentanglement of the typically linked features of material and geometry for an expansive design space.

To demonstrate applications of this workflow, the gradient design tool was fed from the output of topology optimization solvers such as Millipede to demonstrate how this workflow could feasibly create an optimized material distribution.

Figure6_application of stiffness patterning_cs6_edited.jpg

This research focused on creating prototypes with programmable deformations to demonstrate the potential of designing with stiffness gradients. A finite element method (FEM) simulation in SOFiSTiK was performed to verify the observed deformation behavior. The sample’s stiffness gradient input was used to create a mesh with a fine Young’s modulus gradient that approximates the continuous stiffness gradient of the physical object. The simulated mesh showed similar deformation behavior under the application of forces that mirrored the handling of the physical prototype, providing feedback on the distribution of stresses in the deformed sample.