Digital fabrication equipment is increasingly cheap and cheerful- it’s easy to use and easy to own. CNC mills, laser cutters, waterjets- they are just more tools in the toolbox that happen to make custom jobs and one-offs more accessible with their computer-controls. But, unlike the evocative name ‘3D printer’ suggests, they do not enable full control of physical matter. There is no Star-Trek-esque replicator that can automatically build the physical counterpart of an arbitrary digital design (and we’re a ways from getting one).
At the Machines that Make project, we are working on extending our digital fabrication toolbox with equipment that allows us to prototype beyond the confines of the tools we currently have. Why can’t we hook smart subsytems into our lasercutters, or modulate the layer thickness in our extrusion based 3D printers? To provide richer access to tools, we are building modular and open systems of kinematics, sensors/actuators, motion control, virtual machine network interfaces, and CAD/CAM software that can be outfitted with any toolhead. Using these machine-subsystems as building blocks, we can rapidly prototype rapid prototyping machines- for milling/3d printing/scoring/pipetting/writing/coiling/anything. Instead of hacking together existing tools (e.g. attaching a fibreglass-spooling head onto a 3-axis milling machine that you write a weaving toolpath for using rhinocam), we want to elegantly, quickly, and painlessly make machines that make (e.g. outfitting a parametric 3-axis motion stage with a spooling head and writing a software widget to have the stage weave). This talk will outline some of the technical solutions we have built to incorporate the software practices of modularity and reusability into digital fabrication tool design, and how they are making it easier for non-experts to build their own computer-controlled tools.
Some examples of demonstration tools we’ve built so far include tabletop milling machines, 3d printers, scoring machines, and pipetting robots. A live demonstration of assembling and configuring a machine will probably be too slow for the presentation, and just running a prebuilt tool wouldn’t be particularly illustrative, but if there is going to be a demo area during the conference, I can bring any of these machines to show off. You can see some of them here: http://mtm.cba.mit.edu. I’d need a table and power (110V or 220V, less than 300W).
I work on digital fabrication tools, machine communication protocols, and technology for humans.