Moving with probe coordinate system



  • I am wondering if there is already a possibility to make moves with the coordinate system of the probe.

    The issue I have is that for homing I have to move the head to a certain position so I can home because otherwise the Z probe is off the bed.
    But usually the Z probe has an offset to the nozzle.
    That offset is kinda round numbers because I just measured it from the CAD model but the limits of my machine are not because the heads are parked outside the bed so the endstops are there at negative positions compared to the X0Y0 of the bed. So I adjusted the limits that the nozzle is at X0Y0 of the bed which is some not so round numbers.
    When I wanna now position the probe for homing Z I have to manually calculate the position to move to becuase the offset for the probe is just used when doing G30 with coordinates set or in the new grid leveling mechanism.

    Am I again missing some other command?

    A convenient solution would be to consider the probe as T-2 or something because coordinate system translations are normally bound to tools…



  • This is implemented in CNC-for-milling versions of gcode. See G54-G59 here: https://en.wikipedia.org/wiki/G-code

    It's my understanding that for each coordinate-subsystem (G54, G55, G56, etc) the user can set specific offsets from true spindle (tool) zero. In your case, you could define G54 as the Work coordinate system for your probe. You would need to initially set the offset from spindle zero, but then you could just switch to the G54 (or other) coordinate system,. and switch back to default (g53) machine coordinate system when not commanding the probe position specifically. This would allow you to command the actual x/y/ position of the probe while in the corresponding WCS.

    I wonder if using a system like this could be useful for 3d printers? I'm still unclear exactly how the system works, but it seems like a way to manage multiple toolheads and offsets… though we already have t0, t1, etc.... hmm.



  • Nice idea but those commands are not supported by RRF…


  • administrators

    I think you can already define a probe tool. Use M563 to define a tool with no heaters and no drives. Use G10 to define its offsets as those of the probe.

    Please note, G29 and G30 when used with X and Y coordinates already take the probe offset into account, except G30 when doing delta calibration.



  • @dc42:

    I think you can already define a probe tool. Use M563 to define a tool with no heaters and no drives. Use G10 to define its offsets as those of the probe.

    Please note, G29 and G30 when used with X and Y coordinates already take the probe offset into account, except G30 when doing delta calibration.

    Oh yeah… Got it...



  • @Alex9779:

    Nice idea but those commands are not supported by RRF…

    Clearly. I was suggesting ways to implement this that would be compatible with existing gcode standards. But the tool definitions are more applicable to 3d printing and already implemented, so they are a better option clearly. I simply didn't think of defining and selecting a probe tool, which is akin to selecting g54-g59.



  • @bot:

    I simply didn't think of defining and selecting a probe tool

    Ya me neither… Did it now and it works... I defined T99 for that with no scripts. Problems with other scripts when deselecting normal tools won't happen I think because I just wanna use this "probe tool" for positioning the head before homing and stuff when no normal tool has been selected yet...



  • Oh, you could also simply define two macros that G92 in and out of the probe offset, to avoid running the tool change scripts. Edit: nevermind, I don't think relative g92 commands have been implemented yet.


 

Looks like your connection to Duet3D was lost, please wait while we try to reconnect.