From your sketch, i would say you can use the 3p method.
So there is probably something you are doing wrong while using the method, tell us more about how you actually use the method ?
Hi, there is a recent discussion about the same topic, please check
https://www.robot-forum.com/robotforum/fan…l-frame-issues/
Anyway, the three and six point methods basically ESTIMATE the X,Y,Z location of the tip of the tool with respect to the mounting plate using the positional information from the taught points.
IF you can get this data from the cad model, manufacturing drawings or simply by measuring it, just input them directly.
I you want to use the three point method, just teach the thee points to touch a fixed reference
If you can manufacture a calibration tip to use in place of the drill bit would be better.
(Agreeing w/ previous posts)
A pointed pin would be easier to use than a drill bit.
You should be able to manually manipulate X,Y,Z values to see if accuracy gets better/worse.
Life is easier if the bit is straight up and down at robot Zero.
I've already met some used robots, which were totally unable to get the TCP correctly, despite many attempts to define the TCP, remastering, recalibration etc. - probably due to backlash on the gears? Is your robot new?
