I found myself wondering about the basis mechanism of the legs in Theo Jansen's Strandbeest kinetic sculptures and I ran across this site, which provides a measured drawing.
http://letsmakerobots.com/node/5087
I have attached a transcribed copy of the drawing that may be a little more informative and easier to read. Here's a simple legend:
- The red triangles maintain a fixed geometry (i.e. solid plated could be used for the red parts). The fact that they are triangles makes this an easily derived fact
- The blue lines are edges of polygons that deform. Therefore, any vertex with a blue edge must be hinged.
- The circle is the drive for the mechanism. The line "A" shows that the axis of the drive occurs about 16 units above the C,B vertex when the leg is positioned as shown
- The $ and € points are fixed points on the body of the (sic) beest. Therefore the wheel rotates around the $ adn the € forms a "shoulder" joint that bears any load transmitted by the leg.
- The radius M does not appear to be explicit on the diagram Based on scale, it looks to be about 25 units in length. A more precise measurement could be derived using a little trig, maybe. OTOH, my guess is that a ball-park figure is good enough.
I think that this might be easy to print on a 3D printer using the "thin-strip-hinge" technique that Dave used for his printed micro-manipulator. I am going to speculate that it would be fairly easy to build a quadruped with four of these -- front legs 180 degrees out of phase (back legs as well) and the front and back pair offset form each other by 90 degrees.
A leggier version might be a spectacular entry in the Kensington Kinetic Derby.