#linuxcnc-devel Logs

Dec 05 2017

08:47 AM mozmck: I just now saw the discussion about probing from yesterday. Making it configurable is good imo. I have a patch that adds ini settings to disable aborting on probe errors while homing and jogging.
08:49 AM archivist: my probing wish list item is calibration like real CMMs do for the probes
08:51 AM mozmck: My "probe" can be the Z home switch on a plasma table, and obviously when you are homing you don't want it to stop with a probe error.
08:52 AM * jthornton ponders how that could possibly work
08:52 AM mozmck: Works great - been doing it for years!
08:54 AM mozmck: The Z home switch is activated by a floating Z torch holder. Z goes down until torch contacts surface, then continues while floating part slides up until it presses the switch. We can use the same switch for home and probe.
08:55 AM mozmck: of course home in this case is variable depending on metal thickness and all.
09:09 AM rene-dev: archivist that can all be done in user software
09:10 AM rene-dev: archivist you can code some calibration in software
09:10 AM rene-dev: eeh, gcode
10:15 AM jepler: forum will want to reboot, bounce coming soon
10:15 AM seb_kuzm1nsky is now known as seb_kuzminsky
10:24 AM archivist: rene-dev, not really the right place to do that I think
10:56 AM rene-dev: where would you do it?
11:00 AM archivist: with a cmm there will be a table which includes the direction of travel and apparent dia of the sphere of a particular probe tip
11:02 AM archivist: so sets of calibration data that can be invalidated by any change or time
11:06 AM archivist: http://www.renishaw.com/media/thumbnails/320wide/2a7c5d7cbcd943adb9337407e7959c6e.jpg
11:07 AM archivist: that is from http://www.renishaw.com/en/at-the-sharp-end-a-guide-to-cmm-stylus-selection--10927
11:13 AM archivist: a bit like the leadscrew compensation but a circle (would help)/sphere(closer to reality)
11:17 AM rene-dev: yes, but you can do al that in the probing routine
11:17 AM rene-dev: linuxcnc only moves until the probe is tripped. everything else is done in the macro
11:18 AM archivist: currently
05:23 PM Tom_itx is now known as Tom_L
07:58 PM seb_kuzminsky: Re: the mailing list thread about probe behavior: does anyone know how the other controllers behave when a probe trips unexpectedly?
08:11 PM sync: seb_kuzminsky: that depends on the setup
08:12 PM sync: but if you changed the probe into the spindle on a siemens or heidenhain control it will stop instantly if it is tripped
08:12 PM sync: doesn't stop probes being squished tho
08:13 PM sync: as you can hit them off axis
10:03 PM seb_kuzminsky: sync: that's good to know, thanks!