G2, Due+ autolevel/touch plate problem

Home › Forums › TinyG › TinyG Support › G2, Due+ autolevel/touch plate problem

Tagged: 

Viewing 3 posts - 1 through 3 (of 3 total)
  • Author
    Posts
  • March 3, 2017 at 5:18 pm #10295
    niklas
    Member

    I have a Due running with gaups 1.1 shield and I have some troubles getting the touch plate working. All the limit switches working fine so no trouble there but the touch plate is not working.
    I also have a little NO circuit with filtering to make it a little more reliable.

    I first tried to make the probe work with the edge branch, but for some reason the probe seemed to be hardwired on di5 and I wanted it on di6. I tried to switch the inputs on di5 and di6 but always seemed to have the probe on di5. So decided to try out the probing improvements branch, and got a little forward with that.

    Now the trigger seems to be working, but the problems arise when I try to do touch plate detection. The probe goes down and touches the plate, goes upwards, leaves the number 3.75 and G2 goes in to shutdown mode.
    Am I maybe missing something?

    IO active seems to be right since the probe stops working if I invert it. It also does the same thing if the axes are moved manually without power in the motors, so I do not think the problem is in the circuit.

    March 3, 2017 at 7:16 pm #10296
    cmcgrath5035
    Moderator

    We see very little traffic on this Touch Plate here on this forum.
    I have seen quite a bit over at Chilipeppr,

    and also on the Ox forum,

    Even if you have a different machine, the discussion might be helpful.

    Touch Plate is usually considered a CAM layer routine, built on tinyG (G2) primitives.
    There may also be some very specific items at

    related to the di5/di6 issue.

    Good Luck!

    March 5, 2017 at 1:35 pm #10302
    niklas
    Member

    Thanks a lot for the links. Got everything running by reading through the discussions.

    The problem turned out to be that the regular touch plate widget in Chilipeppr uses machine coordinates and I had machine coordinates limited with to 0 on Zmax. When it tried to move upwards, the widget zeroed the Z axis, tried to move upwards from Z0 to outside of the machine limits, and in return g2 went in to shutdown mode.
    This I got past by using the super touch plate widget instead and changing the coordinate system to G54. That way I could use the machine coordinates but still use touchplate leveling for milling.

    The auto-level widget also seems to use machine coordinates. To get a proper probe, I had to zero the Z coodrdinate a few millimeters over the PCB, and also remember to zero the G54 coordinates accordingly. If these were not done properly, chilipeppr sometimes moved downwards instead of upwards after touching the plate, or just went in to shutdown mode. I am glad I was doing the testing with foam under the copperclad board.

    There was also another problem to fix for auto-leveling, sometimes the G2 crashes during process. To fix this, I read on github issues that commenting out colums 120-125 and 135 in gpio.cpp helps. This disables testing the pins with something called “lockout timer”. In the future, I should propbably try to fix the problem instead of disabling it, or maybe disable it only for the probing pin.

  • Author
    Posts
Viewing 3 posts - 1 through 3 (of 3 total)
  • You must be logged in to reply to this topic.