Is there any other explanation for not traveling straight while jogging in WORLD frame besides bad mastering? I had a couple pins slip in one of the main connectors to the robot and it wiped out my count on axes 1 and 2. I mastered to the witness marks and again with a machinist's level, but when I start driving in WORLD frame, I get weird rotation on my TCP and it appears to be drifting in directions it should not be...
Mastering
- droth
- Thread is marked as Resolved.
-
-
What model robot?
-
The only time I had that type of behavior was when I loaded the software for the wrong type of wrist...
Sure it is the same cpu-card ?Else, the mastering is the only thing I know of.
Do you have the mastercounts ? And did you master only axis 1 and 2 ?
Maybe the other axis had bad mastering and you will have to do all the axis. -
Is the TCP correct as well?
-
It is an R-2000iA/R-J3iB. The robot sat for several months, waiting for deployment. Aside from a simple 'exercise' program I ran on it on occasion, the robot is new to us and should be in good working condition. I do not recall this weird movement the last time it was running. Unfortunately, I do not have the master counts, and yes I tried mastering all 6 axes, on 2 separate occasions.
I don't see any active, configured frames, and I'm not sure what you are asking, Tony, but the TCP was my first indicator that something weird was going on. If I try to jog along the Z axis in WORLD coordinates, I get crazy rotation on my TCP. In fact, maybe the tool rotation is all I am seeing and the 'drifting' along X and Y is all in my imagination. I've done more remastering than I care to admit, always using the witness marks on the robot. I've never had a robot react this poorly to mastering, which leads me to wonder if there is something else I might be missing...
-
Was the TCP taught or was it keyed in CAD data?
If this robot has vision I would suggest vision mastering. -
Robot ONLY interpolate linearly (in WORLD, USER or TOOL frames) when the robot is mastered and CALIBRATED
Is not only moving the robot to the marks, in fanucs there is an additional CALIBRATE step
otherwise robot ONLY moves always by joints
Some robots have a special posture for MAster / calibrate
or like in the previous post mentioned, if your robot kinematic does not match, then it move oddlyDisplay the position of the robot and see the coordinates in world frame
if you jog along X, you should only see this coordinate changing and the rest unchanged -
Calibration was carried out, yes. Does a R-2000iA have an odd mastering posture? I just assumed that J2 is vertical and J3 is perpendicular to J2...
Ah. Something just occurred to me. And if this is correct, I'm going to feel really, really dumb. Unfortunately, I have to travel 15 minutes to another facility to check this robot out. I'll let you know what I find...
-
The only rule for mastering a Fanuc is that axis 2 must be at zero to master axis 3 at zero.
Create a zero program. Create a point in the program that is represented in joint and all of the joint angles are zero. Run the program to verify mastering. -
-
The only time I have run into this problem is when I messed up the Tool Center Point.
-
-
It's been nearly 4 years, and I do not recall exactly, but I think I had the robot's posture wrong. J5 specifically. Either I had the wrist pointing straight down and it should have been horizontal or I had it horizontal and it should have been straight down.
But I am 99% sure it was a boneheaded mistake on my part and nothing weird or sinister with the robot.
-
Walker21 When I messed up the tcp it was just because I taught it wrong, or made a program in the wrong tcp for the tool that I was using.
Also, All the Robots I have worked on are for welding. We use a torch alignment tool, really don't bother with the wrist alignment. Did your alignment come out OK after you remastered the axis?
-
Create an account or sign in to comment
You need to be a member in order to leave a comment