Swarf with C-Axis-only movement

[martin_walter]

Hi,

I'm having trouble posting a swarf operation, where I want to machine the outer contour of a simple rectangular plate using the C-Axis. The wall is vertical, which seems to be the "problem".

Postprocessing fails when applying the machine configuration with:

###############################################################################
Error: Error: Tool orientation is not supported for available machine axes.
Error at line: 1

Stack dump:
("Tool orientation is not supported for available machine axes.")@:0

Machine configuration code:

var bAxis = createAxis({ coordinate: 1, table: true, axis: [0, -0.7071067811865475, 0.7071067811865475], range: [-10.0001, 195.0001], preference: 1 });

var cAxis = createAxis({ coordinate: 2, table: true, axis: [0, 0, 1], range: [0, 360], cyclic: true });

machineConfiguration = new MachineConfiguration(bAxis, cAxis);

setMachineConfiguration(machineConfiguration); <-- Post errors out in this line

optimizeMachineAngles2(0);

I found the following posts regarding this subject, but those were not very helpful:

Tool orientation is not supported for available machine axes (4-Axis)

(this is no solution, because of the resulting undercut, and just disabling the B-axis for that operation won't work, because the calculated XY coords won't be correct, then...)

convert-a-absolute-post-processor-to-an-incremental

(this doesn't work for me, because I'm not able to figure out correct values to preset the axes with. I recon that presetting the axes with 0 would be the correct preset, anyway, and that doesn't work. I also tried presetting the axes with values like 45, but that didn't work either.)

One fallback I could think of would be to just use a 2D Contour with a manual NC command that tells me to calculate the C-Axis angle manually for that operation, but I really want to avoid something like that.

Has anyone stumbled upon that problem, and been able to solve it?

And if not, could maybe someone from Autodesk have a look at it?

The IPT is attached, and it seems, that this problem is reproducible with any post using a multiaxis machine configuration.

Cheers, 

Martin

[ArjanDijk · Accepted solution]

Pretty weird Martin. Its not your post settings. Every 5 axis post I tested fails. Also weird because there is not a single point in the operation where it outputs a rotation of any axis if you check with the DUMP post. I think Achim of @GeorgeRoberts should take a look

I think this topic talks about the same thing you want. UYse a contour with the C axis.: https://forums.autodesk.com/t5/hsm-post-processor-forum/haas-umc750-configured-to-rotate-the-c-axis/...

Hope it helps.

[martin_walter]

Hi Arjan,

thanks for having a look at it 🙂

Yep, I found that pretty strange, too. I tried most of the posts I've written so far, and none worked...

I just can't figure out, why this happens.

Could you maybe repost the link? Doesn't seem to work for me.

Best regards,

Martin

[martin_walter]

Hi Arjan,

I think I'll settle for "doesn't work in HSM" 🙂

What I want to do is exactly what keith.clausen was trying to achieve. So I'll go ahead and calculate the angles manually for the 2D contour.

Thanks again!

Regards,

Martin

[martin_walter]

Hi,

could not get it out of my head... 

I persuaded Flow and Multi-Axis Contour operations to do the same. Also with the same result.

This really seems to be a general problem with 5x operations where the normal is parallel to (0,0,1)  

[bob.schultz]

Hello Martin,

The rotary axes in the post processor are controlled by the tool axis vector and in this case the tool axis vector is 0, 0, 1, which equates to a rotation of 0 degrees for the rotary axes.  The post processor does not care what machining operation you used to create the tool path.

What you are trying to accomplish is commonly referred to as Polar Interpolation, where the C-axis will replace one of the linear axis.

We do not have this implemented in the Mill posts, but it is in every Mill/Turn post.  In these posts we refer to it as XZC-mode, since Polar Interpolation is handled in the control.  You could "borrow" this code from a Mill/Turn post and add it to your mill post, but it is more complicated than can be explained in full here.  If you would like to tackle this, please give it a try, feeling free to ask for any assistance that you may need.

[ArjanDijk]

@bob.schultz . You are completely right about the polar interpolation, but what do you think about the post error in 5 axis operations that do not rotate?

[bob.schultz · Accepted solution]

@ArjanDijk  @martin_walter This issue was previously reported and has been resolved.  It should be in the next release of Inventor.