Improvements on Jerk-Accel code #660
Conversation
…eration calculation
…epper loops, adjusted jerk ramps to grblhal calculations in mm instead of time based calculations. now works with low and very low jerk values. known issue: decel ramp does not hit 0 speed at end.
Thanks to @Engigneer, @fpq473 for helping figuring this out!
Can someone create a Wiki page for this? I can give write access for that. |
…t) and new HAL entry point for querying/resetting driver status. Added driver capability flags and settings, $742 - $745, for enabling and inverting motor warning and fault input signals (per axis). Improved jerk acceleration. Ref. PR #660.
i can do that if you want me to explain how the implementation works and how changing the settings affect it? can i use pictures to visualize the points? |
Yes that would be nice. I have opened access now.
Sure, there is an images directory that I use for that. |
Good Morning, here's the first round of improvements on the jerk acceleration code.
now works with a wider band of jerk settings, recommended is still at least 10x accel as a starting point. jogging issues have been diminished greatly but might still crop up as taking a long time to finish the jogging move or having an excessive last segment decel for tiny movement blocks
thanks to: @andrewmarles and @mintracer @engigeer @MBaadsgaard @fpq437 on the printNC discord for troubleshooting, testing and theorycrafting coding help
Note:
Could you mark the jerk setting in the webbuilder as actively beta and maybe expose the ACCERLERATIONTICKS_PER_SECOND and SEGMENT_BUFFER option to people that are choosing it? those are kind of necessary to finetune the system