Jimmy Etienne, et al.
Since we don't do a lot of digital fab...
Additive
Subtractive
Computer Numerically Controlled = CNC
Generally a simple concept, FDM deposits material via CNC machineary
Generally a simple concept, FDM deposits material via CNC machineary
Anatomy
Hotend
Slicing
Addresses a weakness of FDM printers
Addresses a weakness of FDM printers
High level approach
Tetrahedralize
Optimization
Toolpath Generation
Post Processing
Tetrahedralization
Provides a means of representing \(\mathcal{M}\)
\(\Omega\) - Mesh
\(\Gamma\) - Tetrahedral mesh (\(\Gamma_I\) inside \(\Gamma_O\) and out)
\(\mathcal{F}\) - Faces on mesh
\(\forall p \in V, \mathcal{M}(p) = h\) (height)
Layer thickness
Minimum and maximum layer thicknesses
\(\tau_{min} = 0.1d\), \(\tau_{max} =0.75d\)
For 0.4mm nozzle, this gives [0.04mm, 0.3mm], depending on calibration
Slope
Slice at \(\tau_{max}\) in \(\mathcal{M}(\Omega)\), optimize \(\mathcal{M}\) to flatten as many faces as possible
Flatten which faces?
\(\overline{\mathcal{F}}\) - other faces
\(\underline{\mathcal{F}}\) - faces to be flattened
Define an energy function to minimize
\(\lambda\)s are user-chosen trade-off parameters
\(h^0=z\)
Define an energy function to minimize
Attempts to pull vertices of a triangle to the same height
Define an energy function to minimize
\(\mathcal{C}(\underline{\mathcal{F}})\) is the set of connected components of the set to flatten
Define an energy function to minimize
Define an energy function to minimize
Attempts to pull vertices of a triangle to the same height
Constraints on optimization
Thickness constraint
\(\forall t\in \Gamma_I\)
Slope constraint
\(\forall t\in \Gamma_I\)
What if flattened surfaces aren't flat?
What if flattened surfaces aren't flat?
Extrusion rate must be modified in \(\mathcal{T}\), but is available from the gradient of \(\mathcal{M}\)
Error Comparison