UNSOLVED Ideal Acceleration depend on Instant Speed Setting?



  • I've been doing octagon shaped vase tests with staged increases in acceleration and instant speed in separate tests. I'm beginning to be concerned that the ideal acceleration setting changes depending on the current jerk setting and visa versa. Particularly with trying to avoid resonant frequencies on in the constant speed sections of the print.

    Is this fair? Do I really need to consider a full factorial (or Taguchi etc...) to find the best settings of acceleration and jerk?

    I'm 90% certain my main issue is poor motor choice (salvaged steppers, others on order!) but wanted to air these thoughts too.

    Please feel free to just link other thread if it has already been discussed at length.

    This is on Duet v0.6 and 0.8.5 machines. I'm hoping a few Duet 2 Ethernets may become available as people upgrade to Duet3s, and I'll upgrade as soon as finds allow. That said I'm interested in finding the limits of the Dv0.6 before upgrading. I think lack of 'DAA' is likely to be one of them!


  • administrators

    RRF on Duet 0.6 does support DAA. But high jerk settings render DAA ineffective.



  • @dc42 what classes as high jerk? I think I'm down to about 300 on one axis, but compensating with a print speed of 85.

    I think most issues are motors. New ones land Monday. I think most of this work boils down to trying to avoid a resonance right in the middle of the normal print speeds.


Log in to reply