Five Axis Alignment Visual Test Program - Need Help

[Ez_DuzIt123]

Hopefully I can get some help with this one... I have never written a five axis program by hand so forgive me if I am totally off with what I currently have.

Basically what I am trying to do is put a 19MM tooling ball onto the table and then bring a coaxial indicator on it, manually center it, then run the program which rotates the C axis and the B axis so I can visually see the deviations if any exist. To me this seems like a fast way to see if the COR is off without machining a test part.

With the B axis at 0 degrees the program is easy, obviously. But how in the world do I get the B to 90 degrees while the indicator follows the outermost part of the ball? This is the current program I have (Forgive my OCD with repeating modals):

G59 X0.Y0.
S700 M03
G43.4 H100
G1 X0. Y0. Z0. B0. C0. F100.
G1 X0. Y0. Z0. B0. C90. 
G1 X0. Y0. Z0. B0. C180. 
G1 X0. Y0. Z0. B0. C270. 
G1 G17 X0. Y0. Z0. B0. C0. 
[B]G3 G18 X.374 Z-.374 B90. C0. R.374[/B]
[B]G3 G17 X-.374 Y0. Z-.374 B90. C180. R.37[/B]
G49 G1 G17 G90
M5

When I ran this program the indicator started jumping real bad as soon as the B started to rotate so I know this is not correct. Is this even possible to do with only a few lines of code?? I programmed it within Esprit to see if I could use that code but our post processor is not setup right to optimize the toolpath with any arcs so the code I get is a bunch of intermediate points.

 

[thesidetalker]

You don't need arcs.

Setup your work offset to the center of the ball.

Turn on TCPC, go to X0Y0Z0 (with proper 'tool' length of course) and just G1 to different B/C angles.

 

[Ez_DuzIt123]

You don't need arcs.

Setup your work offset to the center of the ball.

Turn on TCPC, go to X0Y0Z0 (with proper 'tool' length of course) and just G1 to different B/C angles.

I think you are correct... I am not sure why I overthought it. I will test this out tomorrow. But what if I was using a Haimer 3D gauge? Then this method would not work. So next question is how would the code look for that??

 

[5 axis Fidia guy]

Easiest thing would be to center the ball, set zero, turn on TCPC (or what ever your tool center point on your control is called) then rotate the table 90 degrees and check again, turn 90 check, etc...

 

[thesidetalker]

I think you are correct... I am not sure why I overthought it. I will test this out tomorrow. But what if I was using a Haimer 3D gauge? Then this method would not work. So next question is how would the code look for that??

It should be the same method, no matter where on the ball you place the indicator tip.

XYZ zero should be center of the ball and tool length offset set to bring the indicator in to touch, whether it be only Z-offset touching the center of the top of the ball, or if your control allows you could apply an XY tool offset to touch off the side as well.

Never tried the XY offset myself. Although thinking about it now I might give it a go as my matsuura supports this I believe. I've only ever done this indicator 'trick' on my haas umc.