Re: Digitizing arm - CMM

On Thu, 13 May 1999 12:35:29 -0300, James P Crombie
<jpcrombie@...> wrote:

>...I am a machinist with lots of available machines to work with and
>own a copy of Rhinocad. Aside from the mechanical part of the
>arm you are going to need an interface box that can convert the signals
>of the 4-6 encoders and either store the points for export to rhino or
>have a little program that will introduce the points to rhino as they are taken.

I would think the best approach would be for someone like Andrew Werby,
who knows the Rhino3d guys really well (I would expect), to ask them
what would be involved in us just being compatible with one of the
existing digitizer's interface, if he'd be willing to do that for us. In
fact, I think somewhere in my Rhino beta testing group archives, there's
a thread about doing this very thing, and IIRC, the Rhino guys said
they'd be willing to help out in such a DIY digitizer interface.

Gotta go do some digging in my notes on that, but as I alluded to above,
I think the easiest path is getting whatever interface specs we'd need
to be compatible, from the Rhino guys directly. The applicable joke in
the trade then is, "It's then just a small-matter-of-programming". B)

IF the std. interface IS in the form of an already-computed point coord,
then that would require just a wee micro with a decent floating point
library, I'd imagine. Probly the most complicated part would be the
calibration routine you'd need to embed in this controller.

Just a few fractoid thots along the way. At any rate, nuthin so far
strikes me as being terribly difficult, time-consuming, nor spendy. And
certainly the code and interface (assuming via serial or parallel port)
to the Arm could be tested with a minimum of hardware and software, in a
PC to begin with, to sanity test and study feasibility.

Gar