Thread Rating:
  • 0 Vote(s) - 0 Average
  • 1
  • 2
  • 3
  • 4
  • 5

Fanuc Rj3ib - RDK Points aren't saving joint values correctly

#1
Good afternoon,

I've run into a quite aggravating issue that I would greatly appreciate some assistance with. In my work cell I have two points, Home and Near Home. Near Home is often used to position the robot for a tool change operation or after a tool change operation to return home at the end of a program. Sometimes I need the robot to move further away from the Near Home position so that it can swing down and rotate without hitting anything. It seems that regardless of what I do, moving the robot with Shift+ALT with either the TCP or the base point on the wrist causes the robot position to snap to something else entirely. This also occurs when I manually jog the robot via the jog menu into the position I need. I've attached a file to help demonstrate this issue. 

In the file, you'll find a program named T7563Z. You'll see that I move to Near Home and then try to simply to -Z with respect to the TCP. For some reason the robot decides to rotate and smash the elbow into the frame of the booth rather than simply backing up from the Near Home position. You can also see strange behavior by moving the robot to Near Home and then trying to move the robot via TCP or wrist.

The images attached demonstrate the strange behavior - the first image sees the robot at the Near Home position. I move the robot +X wrt. the wrist and the robot immediately snaps into the position seen in the second image. 

       


Attached Files
.rdk   rdk_help.rdk (Size: 5.35 MB / Downloads: 86)
#2
There is something weird going on with the program T7563Z. The robot joints should not jump as we see with the simulation.

I'm moving this to the bugs section for now and we'll try to find a solution in the next week or so.
#3
(12-05-2023, 08:37 AM)Albert Wrote: There is something weird going on with the program T7563Z. The robot joints should not jump as we see with the simulation.

I'm moving this to the bugs section for now and we'll try to find a solution in the next week or so.

Appreciate the quick response as always Albert. I had a feeling that it might have something to do with my J2+J3 linked axis joint chart but maybe not.
#4
This issue has been fixed (RoboDK Windows version). The main issue was that RoboDK was unable to calculate the inverse kinematics solution with the elbow down.

You should update RoboDK to the latest version and it should now work properly.
  




Users browsing this thread:
1 Guest(s)