Re: [CAD_CAM_EDM_DRO] Digitizing arms??

doug98105 wrote:

>Group,
>
>Looking for some ideas on making a two axis digitizing arm.
>
>Back in the days of Apple II computers I saw a couple of crude
>digitizing arms built using potentiometers and inputting through the
>game controller input. Position was determined by the resistance
>readings.
>
>With high resolution encoder devices it seems like a much more
>accurate digitizer could be made.
>
>Any thoughts?
>
>

Yeah, look at Faro for how they did it. The problem with these things
is that certain arm positions lead to poor accuracy, because a very small
movement of one joint makes a big change in position.

No doubt a 1000 cycle/rev, or even 2500 cycle/rev encoder would be
a HUGE improvement over the ancient Apple digitizer with pots.
Unless they used special pots, you couldn't possibly get better than
1% angular accuracy, and that is 3.6 degrees! Yikes!

I worked for a medical computer company in the WAY distant past,
and they made "Rho - Theta" and "Phi-Theta" digitizing arms using
very expensive mil-spec rotary and linear pots, and had all sorts of
trouble getting enough resolution with a 12-bit ADC to digitize
the location of a patient's tumors, etc. Linearity was no problem, as
the software had cal tables built for the individual digitizing arm.

Jon