Making simple CNC machines with cardboard!?
http://3dprinting.com/3dprinters/gestalt-modular-machines-make/
http://3dprinting.com/3dprinters/gestalt-modular-machines-make
The only real issue I really have with this project is that they decided to move away from Gcode for a control language. Their “plan” is that Gcode is antiquated and everyone should learn to program their machines with python. There are whole batches of programs available to generate Gcode. Mills, lathes, 3d printers all typically make use of Gcode. Most of those machines have their programs written in Gcode by another piece of software. I forgot to Menton that their “antiquated” language is a Standard.
@Kyle_Kerr It could be both advantage and problem. We always need more options. One other important factor is that the Python is very often used as a teaching language for the beginners. They could directly access machines based on this.
@Kyle_Kerr Sure, but why double the effort? Especially in young age? Also, I think that Gcode can handle some competition 
Change is bad…unless it is great. Trying to make a 3d printer or CNC machine–out of cardboard–while simultaneously killing off g-code sounds like the latter point is trying to reinvent the wheel for no good reason. There is a non-trivial amount of investment in the software mechanics of these machines that would be foolish to try and re-invent solely because “Python”. I’m not disparaging Python. But the author does not seem to have a firm grasp of what is involved in current CNC/3DP firmware or the mechanical design goals therein. But, that doesn’t mean they shouldn’t go down this road, because there could be some great uses for rapid prototyping of the machine itself, which is what I see here. But trying to unseat an established and proven architecture solely for rapid iteration is a weak argument. There is room for both. They don’t want to support g-code? Fine. There are lots of other cool things that could be created with this. Just don’t list replacing CNCs as the primary goal.
Surely it’s just for prototyping?
You can’t make anything that cuts anything resistant with that setup, and the stepper controller will be limited in the kinds of steppers it can drive. With an A4982, that will be about 2A max.
I don’t understand the reasoning behind the cardboard stuff but why not? With a little fiberglass and epoxy it could be pretty slick though I would prefer other materials…
A quick reading of the thesis behind it leaves me unimpressed. Especially their statement of “One of the areas still lacking in Gestalt is the ability to control high speed motion along detailed paths.” Kinda seems like the whole point to me.
As to the replacement of g-code, they seem not to understand their own reference to being “Touring complete.” Don’t like working with g-code? Fine. You don’t have to look at it. Wrap it in functions. Generate it with other programs. The whole idea is to have the computer do that low level stuff for you. I guess they have not figured out what conversational mode is for or how parametric modeling works.
I did enjoy looking at their designs though. Must be nice to have waterjet available for design testing. I sure didn’t see any innovation though. Wow, a Jacquard Loom? I guess that they don’t realize that is where a lot of computer concepts developed?
It looks like a graduate thesis from 2013 that hasn’t had much recent activity, and didn’t really go anywhere since then. Are there recent developments, or is this a dead effort?
Other than replacing a couple items, the github repo seems to have had no real activity since original upload.