[Albert]

The header of the program depends on the post processor and how you customize it. You should customize the header or the setFrame function of your post processor accordingly.

[HonzaMs]

OK. Just one more thing i don't really understand. The generated program that considers the turntable looks exactly the same as with the rotary table. The only difference is the changing table rotation value (E1). That's nonsense, the movements of the arm are completely different in the simulations. Please see the movement instructions (LIN...)

[Albert]

You can see in the comments above the BASE definition a suggestion to edit the BASE variable using the EK function. This behavior can be customized with the post processor.

Another way to customize how you output the BASE variable for KUKA controllers is by setting the BASE_INDEX variable to True:

Code:

# Use the Frame index instead of the pose when the index is provided
# Example: Set to True and rename your reference to "Frame 2" to use BASE_DATA[2])
BASE_INDEX = False

This will link your BASE to the corresponding BASE_DATA index. Fore example, if you use a coordinate system called Frame 4 it will output:

Code:

$BASE = BASE_DATA[4]

[HonzaMs]

I understand. Just please take a look at the recording of the program when the robot works with a turntable and when it works without a turntable. The generated movements in the .src program are the same. Only was added the value of turning the table.
But it can be seen in the simulation that the movements of the arm are different, so I would expect this to be reflected in the generated program as well.
Simulation in RoboDK

[Albert]

I don't agree with your observation. If you check the E1 value, it moves with one program but not the other. Also, the ABC orientation angles are different.

[HonzaMs]

However, the value of "A" has very large changes during the program. If you look at the simulation in the version with a turntable, the spindle is almost always in one position, and in the program I see very large changes in the "A" value.

[Albert]

This is normal. The ABC angles are in the Euler notation and even a small change in orientation can result into large changes in the ABC angles. You can learn more about Euler angles here:
https://robodk.com/blog/robot-euler-angles/

And in the RoboDK documentation:
https://robodk.com/doc/en/Basic-Guide.html#RefFrames

[HonzaMs]

Thanks. I think I already have something wrong with the reference frame setup in roboDK. That's why it generates the header so strangely. I'm trying to find some similar simple turntable program to see if I even have the station tree right.

[Albert]

The station tree looks right. I noticed you have a 200 mm offset along the Z axis to setup your reference frame. You can customize it by double clicking on Machining Origin.