[Quebecois_Sti]

Hello,

我创建一个曲线follow project then exported the generated programs for
testing plasma cutting on a FANUC CRX-10iA.

During the dry-run (before real cutting) on the web side of a small I-Beam,
I am seeing jerky movements because the robot stops at some points during
the small radius/curve of the rectangle. All generated trajectories seems to
be in CNT1 which imply being like a FINE position. The simulation isn't showing
those stops.

Is there a way to avoid stopping movements ? Is the machining project can solve this ?

看视频:https://photos.app.goo.gl/Qu42bzBeoDN9ubgAA

See photo & rdk station attached

Regards,

[AKL]

If you set the rounding to 50 it will change this from "1" CNT1 to CNT50 etc 100 to CNT100. Keep an eye on those fast move speeds 1000m/sec is fast. Can a cobot go this quick?

I do this on a FANUC and just the E and M codes to trigger the plasma on and off well laser for cutting but same idea. Guess on smaller projects could import the cutting program manually if you don't want to code it in the post for testing. I always get the robot to do what I want before I rewrite the post.

Just my 2cent GL

[Quebecois_Sti]

If I change the rounding value to larger CNT value, the robot
isn't "forced" to follow the path, the real positions can be quite
offset of the real one needed...

the faster speeds shown aren't for cutting. they were generated by DK
and according to the tests done, the robot isn't reaching those speed
because of collaborative safety.

[Jeremy]

I would recommend taking a look at video 10-8
https://drive.google.com/drive/folders/1...CCTdy-p7GW

The goal would be to increase the CNT just enough to reach the proper balance between fluid motion and precision.
It's true the robot isn't forced to follow the path with a larger CNT, but if the points are following each other, the discrepancy won't be that noticeable.
Interpolation will be much more visible if you have, let's say 3 points creating a 90 deg angle than if you have 3 points creating a 10 deg angle. `

Jeremy