Re: [CAD_CAM_EDM_DRO] Real-Time measurement

Lee Wenger wrote:

>
> I like tossing out an idea every now and then so I can get battered

That happens to all of us!

> Is there any (Cost Effective?) way to measure the current
> location of the cutter?

If there is, I'd like to know about it.

Instead of real-time while cutting, I have an alternate scheme
for mapping out the screws in 3D (with the spindle off and not
cutting). You'd do this initially after building the machine,
then perhaps every year or season depending on your requirements.

Have a look in the files section, I had posted that a wire be
attached to the spindle tip, and an encoder is used to sense the
amount of wire that is reeled out to go to any reachable 3D location.

In the drawing, I have the end of the wire fixed on the spindle,
and reeling out from 0,0,0. However, I think it might be better to
affix the end of the wire to 0,0,0 and reel it out from the spindle.

These measurements can then be used to not only calibrate the
leadscrews (by travelling down each axis independently and mapping
the distance of the reeled out wire vs the theoretical distance you
have traveled by emitting so many steps), but perhaps also to compensate
for the non-orthoginal errors from imperfect construction of the machine.

ie the distance reeled out is D = sqrt ( x*x + y*y + z*z) and if your
D does not agree with your X,Y,&Z then something is off, and of course
D must be correct (you might want to account for a slight excess reeled
out due to the wire's catenary).

In fact, I want to build a prismatic table (ie X & Y at maybe 60
degrees instead of 90 degrees to each other) and the Z maybe off at
a wild 15 degrees. After compensation, it should be able to machine
a more-or-less perfect cube to single-step resolution, which should
prove the theory and the software.

To get back to your question, the Cost would be some wire, a pulley
and a rotary encoder (and some programming time).

Doug Fortune
http://www.cncKITS.com NEW -> 1 KW power supply kits














.