Re: A "simple" Hexapod project - envelope

This is were the first bit of thought comes into the design. You
have an idea of the working envelope that you want, so you decide on
a design, in my case I like the Triaglide idea, so now what? You
have to do the inverse kinematics, get the cost of components and
set up a formula for economics of the design. This will yield
optimum lengths for legs, screw threads and the angles needed at the
joints and give you the basic geometries that you need to work to.
Keep in mind you can end up with singularities, this is the case
where legs all end up in the same plane or the the table inverts
among other things. So now you have your geometry, you need to do an
error budget i.t.o. stiffness and flexure. For a hexapod this is
fairly easy as there can only be tension or compression in the legs.
This is why i do not like SCARA arm, as there is a lot of force
moments leading to flexure, this means you have to use thicker
parts, this increases mass, etc. etc.

A little mind experiment helps here, the triangle is the only stable
configuration in nature, so for a machine to be rigid in 3D you need
to have at least 3 projected triangles into the x,y and z planes

Two program can actually give you these equations from a drawing,
IIRC Maya is one, the other is Autodesk Inventor or 3d Studio Max

Stephan

--- In CAD_CAM_EDM_DRO@yahoogroups.com, "Graham Stabler"
<eexgs@n...> wrote:

> Thanks again Carl, I don't think I realised how tricky the process
> might be when I first started thinking about the problem. It is

all

> rather alien, we are not used to having virtual models of machines
> just to make them move. I don't mind the maths or the programming
> but is does induce fear!
>
> Graham