[Elie_Roy]

Hi,

I'm running into an issue where a manually-taught reference frame is not being used correctly when generating a program, and I've isolated it down to something that only happens when the robot is mounted on a linear rail (external axis / synchronized mechanism).

Setup:

• Robot: Fanuc R-2000iC/210L, mounted on a linear track (RoboDK mechanism "My Mechanism", external axis E1)
  
• Post processor: stock Fanuc_R30i post
  
• Reference frame: Frame Wall 1, taught and set correctly in the station
  

The problem:
In simulation, a target is correctly reached at X=2000mm, Y=-2000mm in Frame Wall 1's local coordinates. But when I generate the program, the output writes this same point as X=6650mm, Y=-3480mm under UF:1
— which is wrong for Frame Wall 1.
I traced this and confirmed the difference (X: +4650, Y: -1480, Z: +120) exactly matches the offset between my Station ORIGIN and Frame Wall 1. In other words, the generated program is outputting the target's coordinates relative to the station origin, not relative to Frame Wall 1 — even though Frame Wall 1 is explicitly set as the active reference frame before the moves in question.

I've tried all the combinaison of option in the program's option setting. As well as changing the post USE ROBOD_UFRAME = True / False in post processor. nothing works

Thanks,

Élie

[Elie_Roy]

I think I found the bug, The Virtual Assistant plug-in is really good, helped me debug way faster than the other LLMs

The curve, machining project and program MUST be children to the used frame at creation for it to work properly. Thats all.

Élie

[Albert]

Perfect. Thank you for letting us know. We are glad to hear you like our Virtual Assistant.