Quote
A2
@ Solidus Labs
21. Zig Zaging Up Down:
Could the error be coming from the rotation at shoulder pivots?
Maybe there is a singularity when the arm has to rotate quickly?
I showed the math to some one, and they questioned the use of acos.
They didn't explain their reasoning.
Is there a singularity at the center of the build table?
If there is, and it requires significant force to overcome, maybe the motor is skipping a step?
I'm just throwing some ideas out there...
21. I think the zig zagging is caused from the calibrating testing I was doing. I was lowering the number of steps per degree to prevent the print head from ramming into the print bed. Because of the geometry, very small changes in some of the lengths can impact the movement in some very unexpected ways. This is making calibrating very difficult. The play in the z axis on the print head is caused by the rotation on the shoulder pivots. The play is very bad around the edges of the print bed where the arm angle is small and the horizontal pivots are close together. I added a spring to mitigate this but I think it is making movements nonlinear.
Since the worm drive is controlling the angle of the elbow between the two arms we need to solve for that angle from a SSS triangle and that formula is
A =acos( (b^2 + c^2 - a^2)/2bc )
I have been working on implementing the 49 point auto level algorithm that is built into the firmware I am using. The g29 works but after the probing the head sky rockets up on the first move. Anyone have any ideas what causes this? I also discovered that the second effector was also melted from earlier tests