Mechanical Engineers: Questions about Theo Jansen's Mechanism

Factual Questions
1

I’m sure many of you know about Theo Jansen’s amazing mechanical linkage and his application of it through kinetic art pieces.

Here is a flash animation that lets you play around with the various dimensions in the mechanism.

As well, here is a diagram provided by Jansen giving the ideal dimensions. It’s in Dutch but it’s easy to figure out.

I have a few questions about all of this. In the flash animation, isn’t the pink value irrelevant as the involved triangles remain rigid and the parallelogram never loses its defining quality?

Also, in playing with the flash animation I’ve found settings that grant very straight foot paths and clean foot lifts and placements where those settings don’t really resemble the exact values provided in the diagram. Surely the man knows what he is doing and we can assume that the values he provided are optimum. Why do the values he provided differ so greatly from those I’ve generated with the flash animation? Is it because his are derived from real-world scenarios that account for weight shift, structural integrity, leverage, foot slide and speed, body/leg flex, etc?

Thanks to all. I want to build a pedal powered walker and use it to pick up my groceries!

2

Awwww Hamburgers…:frowning:

Nobobody is interested in this stuff? I thought it a very elegant and intriguing device.

Bumpity bump bump.:slight_smile:

3

Looks like a lot of fun. IANAME.

I would agree that the pink value doesn’t change the trace of the legs, but it probabably affects the strength and stability of the structure, which isn’t taken into account in this simulation.

Don’t know about your second question, I didn’t spend enough time with his diagram to try to figure it out.

4

In his diagram, the center of the crankshaft is raised off of the body of the robot, whereas on the animation it is placed directly on the body line.

5

It is some fun. Is it supposed to walk? Otherwise, it’s got a symmetry that makes half of it a duplicate of the other half.

There is some kind of Fourier transform of this thing that would make it clear what lengths to use in some pieces to make the motions, for example, nearly straight lines. You can see some of this in transforms of triangle waves, for example.

6

A tangent question. If these items were powered by a measurable source, say an electric battery, would one be more or less efficient than a rolling machine of equivalent weight? What order of magnitude difference would exist? I’d imagine with absorption mechanisms, a bit of energy could be recovered and converted back into kinetic energy as it walked, but I spent most of physics class in punching contests.

7

With a smooth hard surface, it’s hard to imagine this wouldn’t be less efficient than a rolling machine.

  1. Most importantly, the legs are constantly moving, then slowing/stopping, then moving again. Capturing and re-using the energy from the leg’s motion is going to be difficult and certainly will not be perfect.
  2. Obviously all those joints are more opportunities for energy loss (a two-axle vehicle has four joints by comparison; and the wheeled joints are going to be a more constant stress so easier to keep maximally efficient. Not that there’s a whole lot of energy loss in a good axle joint).
  3. I also suspect that there’s going to be some energy loss from deformation of the limbs; I would think a wheeled vehicle can be more rigid. (I believe most of the non-aerodynamic energy loss in a typical current wheeled vehicle is from flexing the tires).

Obviously, if the surface is less even, and sizes are constrained so that the walker can step over bumps that the wheeled vehicle would have to drive over, the walker gets an advantage, but in general you’re probably not beating wheels on a smooth surface.