the Carpentry Way: Computer Numeric Re-Considerations                                                          

Computer Numeric Re-Considerations

I've been in touch with an architect from Pittsburgh in the past couple of months who has invited me to collaborate on a residential project in urban Pittsburgh. The structure is to be 30' x 60', two main floors with smaller rooms under the roof, which is to be a Mansard. He wants me to design the Mansard, which is to have exposed timber framing, and there is also discussion that I will design a geometrical (helical) main staircase and some intriguing dormers. At this point it remains a discussion so we'll see what happens.

You will note in the above paragraph that I said design, not design and build. The architect's intent, at least with the timber roof portion of the project, is to have the timbers cut by another contractor, using lasers no less. While I remain a bit skeptical at this point, apparently it can be done with larger timbers now, so I'll keep an open mind. The staircase work and the other double curvilinear work with the dormers seems to be intended for production via the world of 5-axis CNC.

So, I am left, in considering the opportunity, assessing the situation here - how do I feel about designing and not building? What is important to me about the process of making objects? Lots of questions have been swirling through my mind of late, and one of the things I have been thinking about is Computer Numeric Control, or CNC for short. I'm already very dedicated to drawing my pieces on the computer, and see many advantages to it, so the steps over to CNC are not so far. A lot to think about however!

My previous experiences looking at CNC-produced timber frame work has not been one of being awestruck. The surface qualities of the timbers I saw retained obvious machining marks, the fit of the joinery was not as good as I would have thought possible, and the product overall had a somewhat 'cold' feel - I know that is a bit of a nebulous description, but it's the best word I could think of to describe it. That said, a tool is a tool, and while I do not believe in the so-called neutrality of technology, I remain open minded when it comes to the capabilities and directions made possible by new forms of technology. At one point, people resisted the introduction of the telephone, while today I hesitate about texting.

Technology can be used in various ways, and not all of those ways can ever be fully envisioned by the developer of a given technology - indeed, there are many examples of a technology being developed for one express purpose which then found wide use in some unexpected field later on.

So, with CNC, what is the purpose - why do companies adopt this technology? From my examination thus far, and from reading various testimonials from companies that have gone to CNC for some aspect of their production, the common reasons include:
CNC finds wide application in production of large multiples of a given part, whether it be elliptical molding, stair-rail parts, plywood chairs, and so forth. CNC is also very useful in prototyping before production begins.

Now, it isn't the mass-production/productivity equation that appeals so much to me. Like any other type of woodworking, if the design is ugly or poorly conceived, CNC adds only a reduction in price. Indeed, lots the CNC-produced stuff I have seen in wood is of little to no appeal to me.

However, that's not the point. The capability of CNC is to produce really any design which can be envisioned, and fabricated, potentially, with high accuracy. That does appeal to me. Here's a point - my interest in bicycles, for instance, involves CNC. My current ride has quite a few parts on it which are produced by CNC, including the headset and hubs, stem, seat post, and so forth. My life rides on those parts - they must be reliable. I like it if they are beautiful too! Actually, the fact that they are produced by CNC is an additional bonus, even a cachet to those products. Look at this handlebar stem for example, made by Italian firm Extralite:
Now, though it may look like a machined chunk of aluminum (which indeed it is!), what you are not seeing is the hundreds of hours of design work that went into this product, including finite element analysis. The stem is machined out to the wazoo and weighs less than 100 grams. I've had it on my bike for several years now with no problems.

Is it cheap? No. But consider what it would cost to have it made by hand. I don't even know if
I'd want it made by hand - the product is pushed to a technical limit of materials, and frankly I think I would trust the CNC more. I certainly know that I could only afford the CNC-produced item.

Consider Karl Holtey, the esteemed British maker of metal-bodied woodworking planes. He's now making a 'transitional plane' (<-- a link), which is wood-bodied with a metal Norris-style adjustment mechanism. The soles of his planes are made of two woods locked together by lots of oblique sliding dovetails:
The above joint is an example of a connection that couldn't be well made any other way than CNC. Intriguingly, he doesn't use his planes to process the surfacing cuts on the plane sole parts - he uses his CNC milling machine.

Take a look at a photo series (<-- link) depicting the steps involved in making his #982 smoothing plane. Anyone out there who would say that these planes are not desirable due to the fact that computer numeric machining is involved? Does it look 'cookie cutter' to anyone out there? My point is that CNC can be used to produce extremely precise objects, and really the beauty is part of the designer's skill. Most people, I think, would choose beautiful and precise over ugly and precise, however there are many who would also choose beautiful and poorly made over ugly and precisely made. Funny how that goes - and that's why there are so many neat designer products that don't last very long. Consumers usually are not oriented to the details of how something is made or care even about how long it will last. This is the consumer society and that is how the conditioning has been done.

I've got more to say on CNC - today's post is the opening salvo about a topic to which I have devoted a fair amount of thought recently. --> on to part II