5 axis kinematics

Controllers Smoothie
1

I’m working on a 5 axis pick and place machine. There is basically a XYZ coordinate and 2 angles that should make the machine move to the requested position. Its ok if the ABC axis show some movement error in timing, as long as they arrive at the good position, before the machine traverse to the next line of code.

From understating this line:

NOTE that the parameters passed into A B or C in a G0 or G1 are raw, they do not go through the arm solution, they are the values that will be passed to the stepper based on steps_per_mm for that axis (although it may be interpreted as steps per degree for instance).

So the ABC axis are not taking part in the arm solution, BUT can I use the ABC value’s in the kinematics section to make the right offsets based on the ABC value’s?

If i understand correctly, it might not make Axis B arrive at the same time as the axis that take part of the arm solution calculations (XYZ), but it would still end up in the proper location.

Is the above correct?

PS, I love to do the maths for the kinematics part, but I’m little blow away from all the code. I can program C/C++, but if at some point someone would like assist that would be great.

Imported from wikidot

2

Saying it doesn’t go through the arm solution just meant it is not concerned by things like delta kinematics.
It is still synchronized thought, they same way extruder axes are for example.

3

That’s great, that will look stunning if in motion!
So you can use XYZ axis for fancy arm solutions, but just not the ABC. But it is still possible to use the ABC parameters in the XYZ calculations, i assume.

for example:
actuator_mm[ALPHA_STEPPER] = fantasyMachine_mm[X_AXIS];

could be:
actuator_mm[ALPHA_STEPPER] = fantasyMachine_mm[X_AXIS] + fantasyMachine_mm(A_AXIS);

But than with a bit more complicated math off course…

4

I have no idea if that’s possible for new arm solutions. It’s not the case of any existing ones.

5

Ok, well would be great, but it could also be pre-calculated, especially as all axis will be synchronized anyways.
Ill do a drawing and add the math, maybe someone would like to look at the code with me from there.