The Incremental Cost of Part Complexity

I was introduced to parametric solids modeling about the time I completed building the pipe organ for Zion Lutheran. The leading program at that time was and still is SolidWorks. Although I wanted to use solids modeling then, I continued to use 2D computer drafting because of the high cost of 3D design programs including SolidWorks.

A few years ago, the cost of a reasonably well featured solids modeling program became low enough that I could justify incorporating it into my design work. Now I finally had a real engineering design program to use. The first thing I immediately noticed was a reduction in errors in the workshop. A good solids design requires that one accurately model each part and assembly, and having those represented in drawing format let me clearly see each part feature, and the relationship of each part to its parent assembly, in three dimensions. All of that helped me to build better in the workshop.

But something else happened that I did not count on. The parts I designed started becoming more complex. Parts began containing more features as parts started taking on more function within an assembly while the assemblies themselves began to have less parts overall. Look at the exploded view of the assembly above as an example. The assembly itself has only four parts. That seems simple enough. But look at the notched rail. That part is designed to hold a component mounted through the hole located along its top. The notches themselves locate the assembly along rails that are part of a larger assembly. And the two vertical slot mortises receive the tenons of two tenoned cross members that hold it all together.

Each of these features, the notches, hole, and mortises, must be accurately and sequentially machined until that one part is completed. The interesting thing I find now is that the cost of producing each feature is not simply the time it takes to machine that feature, but the additive cost of producing each preceding feature if a mistake is made while producing that feature. This is a problem because anyone who has done any woodworking knows that the activity is inherently error prone. A mistake at the end means having to go back and do a lot of work all over again, all the blade changes, tool setups, sizing, etc., and all for one part.

The projects that I've journaled here reflect an iterative change in part design and part to assembly relationship for me. All seem deceptively simple enough, but all are providing me with valuable experience in designing and building within a new paradigm. I've known others who have built in this style of complex efficiency and part economy especially among the German pipe organ builders. I find it an interesting and intelligent way to build.