Re: re:Digitizing

Marcus wrote:

>I have been following the digitizing discussion with interest.
>What I haven't yet seen, is a comment on how to minimize the crash
>potential<s>
>What's wrong with tailoring the probing routine to the general
>geometry that you are trying to more fully define?

Marcus,

Exactly! If you read between the lines of Terry R's post, you'll
recall he mentioned his daughters' company had MANY glass plates with
etched lines (for different objects).

This 'probepath' selection is a VERY important part of digitising
WELL.

I've already outlined the way I think it 'should' be done. Probably
influenced by the fact that the first 'probe' I saw was that "servo
feedback" 'space probe' at Computer Continuum which 'traced the
surface'; rather than probe and repeat.

The "move to a point on a grid at full retract, probe 'til hit,
retract fully, next" routine is like milling a part with the release
plane set at maximum retract! Ridiculous, IMO.

Freds' idea of speeding up the process by the use of an 'external'
logger is interesting, but I think he's gonna have problems
with 'real world' parts...

>Is this how it is done on CMM's?

I've only used the old-generation CMM's which were/are largely manual
affairs...

Hope this helps.

Ballendo

P.S. What about a "servo feedback" probe? Any thoughts?