stepper motors out of sync only in webcontrol



  • Hello,

    So I have a gantry style 3D positioning system that uses a duet 2 for it's controller. A researcher who is no longer with us wrote a matlab script that can controls this setup via commands sent over COM. I wanted to move the system away from that dependicancy but there seems to be a serious problem when using the webcontrol. The X axis as seen by the figure is controlled with two stepper motors in parallel however when jogging from the webcontrol the X traversal can be seen as significantly out of sync when moving greater than 10 mm requiring Estop to prevent damage if it were to go any further.

    This is where it gets weird, when executing commands from COM this problem does not exist. Only in the Webcontrol. Is there something that is not being set up properly that is causing this? I can upload any configuration flies or procedures as needed please let me know.

    Also maybe unrelated but it seems to be very jerky when it stops or moves short distances, is there anything that can be done to mitigate this and what cost in performance is there in doing this? This machine will experience a very low amount of resistance at the tool head.
    alt text


  • administrators

    @mmaslakowski said in stepper motors out of sync only in webcontrol:

    This is where it gets weird, when executing commands from COM this problem does not exist. Only in the Webcontrol. Is there something that is not being set up properly that is causing this? I can upload any configuration flies or procedures as needed please let me know.

    I suspect the web control is commanding movement at higher speeds than your serial port program. If so, the fix is to make sure that the M203 commands in config.g limit the speeds to values that your machine can achieve without skipping steps.



  • @dc42 said in stepper motors out of sync only in webcontrol:

    M203

    So I check the Config File and the value read:

    M203 X15000 Y15000 Z15000 ; Set max speed, 15000mm/min

    This is was off by about a factor of 10 in all axis, must have been a mistake by the operator who did the original setup, still shocked the com execution never in countered this problem.

    Thank you for pointing me to the problem!


Log in to reply