Kelvin disc construction
Rather than construct the main disc using concentric circles, which would restrict the model's scale to what fits onto an A4 page, I decided to go with these radial sections. This approach gives the disc much more structure and stability than concentric circles, the smaller elements are easier to handle, there is less likelihood of warping (please insert Star Trek pun here), and printing out large models is cheap and easy, since one petal per A4 page results in a truly huge model. Even when the parts pages are printed out so that two pages of parts fit onto one sheet of paper, the finished product still measures 59 cm in length overall. The main disadvantage of the radial section/petal approach that I have discovered is that a) it is a fairly monotonous affair, and b) one needs to pay a great deal of attention to maintain proper alignment. Obviously, the more parts there are, the greater potential there is for the disc to go out of true when all of the mistakes in each individual piece are added up. Right now I am experimenting with different radial section angles to see what the ideal proportion is balancing on the one hand the desire to avoid the monotony of repetition and on the other hand the desire to achieve a flat disc. In both cases, of course, I am limited to what I can squeeze onto an A4 page.