# Setting up rotational axis offsets and "rot-points"

15 Jan 2021 00:43 - 15 Jan 2021 00:43 #195383 by kkuehn
I'm having some difficulty understanding how to calibrate the offsets to properly be able to use 5-axis kinematics.
There are six halpins that I believe are related (or that I've seen other people using, especially on the very long switchkins/TCP thread). I'm using an 4th and 5th axis on my mill with the xyac-trt/switchkins kinematics configuration (my machine matches the vismach simulation for xyzac-trt).
setp xyzac-trt-kins.x-offset
setp xyzac-trt-kins.y-offset
setp xyzac-trt-kins.z-offset
setp xyzac-trt-kins.x-rot-point
setp xyzac-trt-kins.y-rot-point
setp xyzac-trt-kins.z-rot-point

The a-axis center of rotation is about 50mm above the top of the x-axis table. My c-axis center of rotation is aligned with the a-axis center of rotation. The top of the c-axis chuck is about 3cm above the a-axis center of rotation.

My question is: what /how do I need to change from these pins in order to calibrate the motion of the machine? I thought that it was only necessary to change the offsets (based on this: linuxcnc.org/docs/2.8/html/motion/5-axis...inematics_components ), but the motion was totally wrong when in TCP mode.

Thanks,
Last edit: 15 Jan 2021 00:43 by kkuehn.

15 Jan 2021 06:05 - 15 Jan 2021 06:08 #195405 by kkuehn
Ok I looked into it more and I think I understand it a bit better. I believe that the rot point is the center of rotation distance relative to the machine home location (set for each axis). If for example, the center of rotation (where the c-axis center of rotation intersects the a-axis center of rotation) is in the middle of the table that is something like 100mm from the home position of the y-axis; would the halpin setp xyzac-trt-kins.y-rot-point then be set to 100mm?

I'll go ahead and keep testing my hypotheses, but if somebody could point me to a document that explains how to actually measure and enter the offsets I would be very grateful.
Last edit: 15 Jan 2021 06:08 by kkuehn.

15 Jan 2021 15:07 #195437 by andypugh
Yes, I would expect these values to be in the machine absolute coordinates.

To measure the values you first need to know where you are measuring _from_. (you are measuring _to_ the rotation axis). This is likely to be fairly obvious for X and Y. It might be ambiguous for Z. It is probably the tool length reference point, and that might be something that you simply have to decide, and then stay consistent to.

01 Mar 2021 20:30 #200728 by JackRay
Good evening hall,
I have the same problem on the Z axis. Concerning the X&Y axes it is necessary to enter the values ​​of the center of the C axis from the origin of the machine.

Example:

HALCMD = setp xyzac-trt-kins.x-rot-point 163.497
HALCMD = setp xyzac-trt-kins.y-rot-point 265.305

Concerning the Z axis I do not understand which value must be entered.
Despite several attempts my movements are wrong

1 / The Z value is it from the origin of Z to the roration axis of the A axis ex:
HALCMD = setp xyzac-trt-kins.z-rot-point -200 and HALCMD = sets: z-offset -50

or

2 / HALCMD = setp xyzac-trt-kins.z-rot-point -250 like the X / Y axes?

Thank you

02 Mar 2021 22:57 #200844 by andypugh
Sorry, I really don't know.

If experimentation doesn't make it clear, you might have to look at the calculations done in the kinematics module.

But jogging the rotary whilst watching the behaviour of the other axes for different offset values should help work it out.

Note that you can set the values in a live system from the command-line while testing.

Your syntax above is a bit off.

halcmd setp xyzac-trt-kins.z-rot-point -200

Is the right syntax.