Has anyone here used their Longmill with a digitizing probe to “scan” an object in to a 3D file that can be edited? I’ve seen a few videos and product examples that can do this and it’s pretty amazing and could be quite useful.
Here is an example of it in action: CNC Digitizing: 3D Digitizer probe for CNC Millings Machines and Routers - YouTube
-Jeff
Jeff, I did this with a simple probe I made up connected to the same pins the zeroing block is connected to with CNCjs and bcnc. The tip f the probe on mine is kind of wide for steep contours, but works well. Be prepared for slow going if you use lots of points for a detailed surface.
Bill you never cease to amaze me! Were the results good and how did you extract the file? Could you load it in to Fusion 360?
-Jeff
The file is a file of coordinates in X, Y, Z format. bcnc and CNCjs format them differently, but it’s completely obvious how they’re done when you look at them.
If you want to use them directly to adjust a cut for an uneven surface, like cutting on a cupped board, each program reads their own and will make a new gcode file with adjusted Z heights. I used them to re-create a surface. I just suck them into Excel, clean them up so they’re uniform and in strict X, Y, Z form, and scale the Zs so they’re all positive. Then bring them into MeshLab and create an STL. You can get a pretty detailed STL, but for complicated surfaces you need LOTS of points, and the time to make all the probes gets long - hours long.
Bill, are you secretly Elon Musk? You’re always 2 steps ahead on this stuff. Thanks for sharing the details! How was the fidelity of the stuff you created? Were you pretty happy with it?
-Jeff
Jeff, the quality of the output depends a lot on the relative surface levels in the part you’re probing, so on parts with gentle hills and subtle elevation changes you can use a wider point spacing than you can on a part with a large highest-to-lowest point elevation and rapid elevation changes. The size of the probe point matters too. With a fat one like the one I used in the video (the one in the pic above) you can’t get the detail you can with a finer point. And you need a long, narrow probe like the prelim version I sent you to do samples with large elevation changes. Remember that doubling the point count in both direction quadruples the time it takes.
Did the drag knife Fusion files work for you?
@BillKorn OK, that the HW, and it all makes sense. But what that about the routine to do the probing? I’ve been thinking about this and I see the gcode commands that would make this possible, but I don’t know enough about the host interface to try to capture the output data. Did you use gSender? If so, can you publish a simple example of how to do this in gSender?
Has anyone attempted this with a rotary?
Welcome Jonas, you’ll surely find anything and everything you need here.
