I am a third year B.Arch student. My ex stepdad invented Ricedream. ( lucasspiegel.com )

 

----

 

End of Term Summary:

 

 

Intro:

 

    The CNC router presents many opportunities for the creation of objects that would otherwise be impossible or overly time consuming. However, it also brings with it a whole set of challenges that can outweigh these opportunities if they are not properly dealt with.

    One thing about the router that takes a while to understand is its exactitude. Having worked with the laser cutter quite a bit I noticed that there is a similar misconception about the CNC router: that because its movements are dictated by a computer, its operation is infinitely exact. This is untrue of both, and even more problematic when applied to the router.  The operations, if properly setup, are extremely exact. The problem is that the bits and the stock are not.  Using expensive prototyping foam, a tiny bit, and a very small step-over, one can approach an exactness found in their computer model, but often the materiality of the stock is important (e.g., the look and feel of wood), using a small bit limits the depth of your cut, and small step-overs result in very time consuming milling operations.  Also, when two-sided milling is required it becomes very difficult to have the top and bottom perfectly aligned along all three axes.  

    Another limit to the router is fact that it rarely outputs finished objects.  There is inevitably more work to do after the milling in order to achieve the finish, assembly and/or casting desired.  This in turn requires that you have expertise in dealing with your material (e.g., woodworking, slip-casting, etc.).  This makes it impractical to use a stock you are not already adept at working with by hand.

 

 

Projects:

 

 

My first project was a bracelet.  My intention was to use the computer-based nature of the process to mill a sophisticated shape that would be impossible for me to carve by hand with any accuracy.  They bracelet is a kind of möbius strip (technically a one-sided object, my bracelet--at least the second iteration--was a two-sided object). It took me two iterations to achieve the scale, surface, and geometry I was happy with.

 

 

the first model:

 

...and here is the full model, with "squewers" to keep all the parts in place:

 

here is the milled piece, before removing it from the stock:

 

...and here are images of the sanded piece:

 

 the second bracelet...

 

    My second project was platter. The initial idea was that I would make a plaster cast of the milled positive, then slip-cast more positives in porcelain. The hope was that I could make a detailed, faceted shape that would be reinforced by the atmospheric surface effects of wood fired ceramics. This turned out to be too large an undertaking to begin half way through the term.  I ended up designing a plater based on a fractal organization of pentagons using the Fibonacci sequence, and milling a small sample out of prototyping foam (would have been good to for casting), and a larger platter out of wood (nicer stock, but harder to finish).

 

 the foam platter...

 

the wooden platter...

 

    My final project was a series of seven nesting dodecahedrons.  I milled them out of a single piece of paduke wood. The router created an interesting layering of textures--the geometry, overlaid by the strips of the milling, and super imposed over the grain of the wood. In order to get them to nest, however, I had to sand smooth their exteriors. After the necessary finishing, they were a more of a functional equivalent of my computer model rather than an exact duplicate.

 

milling the dodecahedrons...

 

the finished objects...

 

they nest or stack...

 

Conclusion:

 

    An important strategy to use is to design for the stock you have. It is much easier in the long run. If you design your model first, you have to either adjust it before milling, or alter the milled piece to make up for the problems your incompatible stock created.

    One must also remember that finishing is just as important, or more so, than the milling itself. Design based on how you will be able to finish the piece. For example, I made some beautiful shapes in my wood platter that were impossible to sand smooth. It would have been much easier if I had stuck to platonic shapes, or larger/simpler geometries.

    The CNC router is such a complex machine--and one that only gets more complicated when combined with the properties of the stock you choose--that it is generally impractical for production.  It is however quite useful for prototyping or the machining of single objects. Because of this complexity it is easy to find yourself with diminishing returns as you balance the power and the challenges inherent to the machine. For this reason, I believe that it would be best used with discretion.  Obviously for this class we needed to jump in and start designing things to see what it could do.  Ideally however, it should be seen as a tool like any other, to be used only when its particular strengths are suited to solving an existing design/fabrication problem. When used with this kind of discretion, and with an understanding of its limits, it is capable of producing compelling and innovative results.