LinuxCNC S-Curve Accelerations

@grandixximo 

Is it possible that if the maximum speed is not reached (without t3/s3), the jerking exceeds the limits?


   

gruß
Rüdiger

I've not noticed this behavior, but I'm not omniscient, I'd need more information, your pictures are lacking context...

@grandixximo 

Attached are the two NC files.

@grandixximo 

In the blue circles, the jerk is outside the limits.
In the red circle, the speed of the axis should, in my opinion, always be the same.

 

Have been working on the S-curve, even when not posting here, YangYang is going to merge Ruckig, will see how mantainers react, it is MIT license, so we should be good, we don't need the pro version, and YangYang can fix it if needed, probably next week update on planner type 1
As for Planner Type 2 progress continues, I am now done with Feed Override phase, doing some cleaning up, and moving to Blending soon, arcs are working from curve9 from Tormach, planner type 2 also has ruckig but in uspace, which makes some things a bit more complex.
Progress is ongoing, thank you all for the support, and the testing, there will be new release to test probably next week.

that's awesome great progress.

quick question

have you caught up with Rob from tormach yet about including any of their work from the 1500mx machines

Had a 1 hour 30min meeting with Rob and YangYang, and Andy Pugh was also there for the first half an hour, it was extremely nice to make both acquaintances.
We discussed a bit of everything, I've asked Rob when he has time to take a look at the 9d branch and give me some direction on it.
We did not talk about 1500mx, but in general, we agree that ideally in our dream TP, Kinematics need be moved first, and be the reference for limits pos-vel-acc-jerk. We agree S-curve early in the planner is more ideal than what Tormach does now, we agree Tormach  SLERP approach does not have continuity for jerk control, Rob warned about Beziers, as they will be a headache for look-ahead. It was very productive meeting in my opinion, I talked maybe too much lol, but it was kind of necessary to have Rob understand what are the problems we face with the current architecture.
Anyhow I pointed him at what has been done so far in the 9d branch, and asked him to give me an honest opinion, since everything is vibe coded, hope it does not do his head in, but I told him to be honest, and if I have to scrap everything and start from 0, I would accept that...
YangYang will still work on improving planner 1, since that is still the most battle tested codebase we have, the 9d branch is an exploratory branch, which will be used to explore different approaches, and see what sticks, I think is a good approach, because out of me playing with Ruckig got YangYang interested in it, and he now moved it in the source tree and is using it heavily to fix the slights spikes we were seeing before.
Ruckig has MIT license, and we also had AI port it to C++ 11 because we want to support old debian 10, we still have machines running on it, which I am not interested in updating to debian 13.
github.com/mika4128/ruckig.git

New PR from YangYang is ready, hopefully will be merged soon in master

This adds ruckig in the planner type 1
if you want to test it get the binary from

github.com/LinuxCNC/linuxcnc/actions/runs/21975584400

or build source from my repo

github.com/grandixximo/linuxcnc

Testers are welcome to report here or in github, thanks!

grandixximo wrote:

New PR from YangYang is ready, hopefully will be merged soon in master

This adds ruckig in the planner type 1
if you want to test it get the binary from

github.com/LinuxCNC/linuxcnc/actions/runs/21975584400

or build source from my repo

github.com/grandixximo/linuxcnc

Testers are welcome to report here or in github, thanks!
 

Hello gentlemen,

movement are incredible smooth ... basic shaping what I have tested till now is pretty ... but during movement at more points during simulation of rotation interpolation its jumping or woobing ... kind off braking and next accelleration 

most spiking during cloud of points are gone ... some spiking are still present but it a huuuuge step forward 

testing setup 2ms benchtop profibus master 3axis geany gui ... 3 servo axis with synchronized speed control loop at 2ms

some testing fault popups triggered during the movement of file 3_1001.ngc

I am still testing at default simulation file from past but splitted to separated operations ..

thanks a lot for your progress ... I am watching it every single day with mount open and box of popcorn to reading the edits ... 

regards E.

My initial tests showed no violations of the limits for jerk and acceleration.

Has the track speed for circles G2/G3 been reduced in the past?

The speed with G2/G3 seems very low to me.

Regards,
Rüdiger