Machine simulation vs without - Posted code differs, and is wrong?

[max2KL2E]

Greetings Autodesk forum,

Please see attached screenshot, my question is why does the posted code change from a working and good Heidenhain vektor program format, to a non working rotary angle format, for simultaneous 5 axis milling, when posted out with the "use machine configuration" ticked on, for the actual machine simulation model.

Both are the same toolpath from fusion 360.

(I know I could get the rotary angles working with enough effort on the machine, but vektor programming format is simply better for a number of reasons I wont get into here)

In the past i have simply unticked "use machine configuration" in the post processing dialogue, this works good, but in my opinion is not a true solution, could anyone shed some light as to why this actually happens?

The end goal is i want fusion to simulate exactly what is going to happen on the machine linking moves and all.

Could somebody maybe provide a timeline of when that is planned for, or if its currently in development, ect.

Heidenhain iTNC 530

Best regards,

Max

[a.laasW8M6T]

I agree the tool-axis vectors are better, but I think that machine simulation needs to use axis angles to work currently.

The problem with using axis angles with Heidenhain is if you have any 3d basic rotation applied to the preset, axis angles will not work correctly, they must be tool vectors.

there is a bit of a difference how this is handled tnc530 vs tnc640 too.

this video covers it well

https://www.youtube.com/watch?v=rYJQTo0Lowo&list=PLDlJ9JmkFDEPKOhZyv3_iqeghOGNFEGj4&index=25 

(I know that you probably know all this, but this is for the benefit of others who may not be familiar)

I don't think you will get an answer on when linking/retract moves will be available, I have asked in the past and its one of those things that autodesk cant commit to giving a timeline for.

[leo.castellon]

That is a great resource! 

LeoC

[licensKKGSY]

As far as I'm aware, vector programming is inherently not compatible with machine simulation, as there might be multiple machine posture solutions for any given programmed vector. The simulation software need to know the rotary solution to that vector in order to give the correct simulation. For vector programming these solution are found by the machine controller, which is also the beautiful simplicity of vector programming.

If simulation/verification wasn't such an important part, using vector would be a no-brainer and post processors would be a lot simpler. Some controllers have their own simulation though. 

[a.laasW8M6T]

I realised now, that I was incorrect about my assertion that you need toolaxis vectors with 3D basic rotation.

This is true if you use M128 for tcp

If you use TCPM(which is better anyway), you can use:

That Axis Spatial takes into account the 3D Basic rotation,

The standard Heidenhain post for fusion is set to use Axis Position so It will not take into account the 3D basic rot.

Of course if you use a 3D basic rotation what the machine does will not match the fusion simulation anyway