Hello,
Following an error message C26 (Motor encoder index drift detected), we had to change a joint of our robot. Our distributor changed it, as well as a simple calibration.
After that, our customer could not find the precision that the robot had before the change.
Our distributor explained to us that the single calibration is not very precise and that it would be necessary to carry out a double calibration, which requires a dismantling of the robot at UR (robot head, robot and test/offset).
Our customer has questions about the robustness of the robot, and its calibration methods.
Have you ever had similar problems?
Is there a simpler calibration system (tooling), more accurate than the simple calibration and which does not require a return of the robot to UR? Do you have any feedback on the calibration methode by Key Waypoints?
Thanks for your answer
M.Baffert
I am biased on this topic. Working on solutions for industrial robot accuracy improvement. We are developing an easy to use cobot calibration method. You can read more on https://cognibotics.com/mptoolsuite/.
Best regards
Magnus
1 Like
Hi Magnus
Thank you for your anwser. I think we have the same idea, but we’ll make an other process:
-We’ll want use a tool for save a new joint calibration before each shipment (out of the factory)
-This tool lock the robot in know position and it enable to read the angular joint position
-When we’ll change a joint, we’ll can use this tool for lock the robot and compare the position robot with the position robot ex-factory.
I’ve probably changed out 30 joints ranging from CB3 to E-series and have never had any customer complain about waypoint accuracy after the procedure. This was most likely a fluke event. Key waypoints is a great new feature they have added. unfortunately I’ve yet to have a customer that has set it up as it’s pretty new. I’m not sure how well documented the procedure is either.
Hi dbarton
Its not realy the waypoint accuracy. There are a offset between the robot before and after the joint remplacement.
If the offset is in the x or y, I would assume when the robot was remounted on the base, it had a bit of play to move. Did you have a dowel pin installed?
Which joint was changed? Do the points get farther off, the further the points are from the base?
Yes, there are 2 dowels pin installed for fix the base mount. We have the problem on 3 robots, one it’s the base, an other it’s the shoulder and the last it’s the first wrist. Yes, the points get farther off, the further the points are from the base