3+2 milling tool/wcs orientation translation to post processor

[benjamin396SN]

Hello,

I have a question about the tool orientation  when milling 3+2 operations.

The machine i am using is a MAZAK Variaxis 200 with MAZATROL 640M PC-FUSION-CNC.

It is a 5 axis machine with A- (+30°;-120°)and C-axis(+360°;-360°) in the table(rotation around X and Z).

The postprocessor is the standard Mazak postprocessor from the library that i modified to 3+2 machining (see attachment).

The image below is the part i want to make with stock and the WCS position and orientation which is the same as the machine WCS orientation.

Stock + (Machine) WCS orientation

The following 4 images shows how i programmed to make a chamfer with a straight mill on both sides. The postprocessor correctly posts it to "A-45 C90" and "A-45 C-90" (see attachment 100.eia). When i postprocess the images on the left, will the machine mill it correctly (wcs/tool orientation is the same as machine wcs orientation).

But when i postprocess the images on the right (wcs/tool orientation is not the same as machine wcs orientation),

will the machine mill it incorrectly (mirror image). The postprocessor outputs exactly the same program when you flip the wcs or not.

Is this normal behaviour that you have to think about the WCS/tool orientation so it will match the WCS orientation of your machine with 3+2 milling? Or is this something that can be fixed in fusion 360 or with adjusting the postprocessor?

Chamfer rechter frees 1Chamfer rechter frees 1 wrong axis rotation

Chamfer rechter frees 2Chamfer rechter frees 2 wrong axis rotation

If this is normal that you have to check the WCS/tool orientation for 3+2 milling, i still have another question.

For the same part i still have to mill a 15° plane on both sides (see images below).

In this case the machine could mill one side with the following positions: "A15 C0" or "A-15 C±180"

And the other side could be: "A-15 C0" or "A15 C±180"

But the postprocessor always outputs the following solution no matter how i rotate the WCS:

"A15 C0" for one side (first image below) and "A15 C-180" for the other side (third image below).

(see attachment 101.eia)

And same story as above if i do not position my WCS/tool orientation correctly the machine will mill it incorrectly (mirror image) => but i have to postprocess first to know how i need to put my wcs/tool orientation...

The question is, is there a way i can tell fusion or adjust the postprocessor so i can choose how the table moves?

Like for example i want to mill these 2 planes with "A15 C0" for the first plane and (first image below) "A-15 C0" for the second plane (second image below).

Schuin vlak 15° 1Schuin vlak 15° 2 wrong axis rotationSchuin vlak 15° 2

P.S. if you are wondering what the following code means in the programs: G65P9900 U0. V0. W0. A15. C0. M54 S59

This is a separate program (9900.eia) that Mazak sent me to translate the machine WCS offset (G54 in this example) to a new WCS offset (G59 in this example) for 3+2 milling.

This is because the machine does not understand G43.4 or G68.2 (see postprocessor).

Thanks in advance

Benjamin

[jahnj0584]

You just have to make sure your X/Y axis are also set properly, and apply either Yes/Reversed to your A/C axis so they actually rotate the right way on the machine. Fusion will post out different codes if they are set wrong.

[DarthBane55 · Accepted solution]

When you just have a part in your file, it is difficult to see if you orient your tool correctly or not.  If you program in an assembly, and have the machine pallet or some other reference, it really helps, you can then figure out that for example, the Y axis should always be away from the machine pallet, or something similar, whatever works for you (Y axis away from machine pallet is correct for A-C machines).  Important to know that once you pick your Z axis, whatever angle your A axis is at, if you pick the machine pallet for your Y axis, the orientation works out correctly...  meaning the Z axis is the "master", and even if your A axis is tilted at A-35º, if you pick the pallet for your Y orientation, you will see that the Y axis will not become perpendicular to the pallet, because of the Z...  so the Y axis will be oriented correctly at A-35, but in relation to the pallet...   if that makes sense...  It's really easy to pick your orientation correctly if you have a reference that you can pick in the file and not just the part floating in space.

[benjamin396SN]

Thank you for the response!

So i understand now that in 3+2 programming you have to orient the tool orientation correctly according to the machine WCS.

But i still cannot recreate how i can choose my A- and C-axis in the second example (last 3 images).

I want to program "A15 C0" in image "Schuin vlak 15° 1", which it posts correctly.

And in image "Schuin vlak 15° 2 wrong axis rotation" i want fusion to post "A-15 C0",

but it posts "A15 C-180" (= correct tool orientation in the last image).

So i think this is what @jahnj0584 means with "apply either Yes/Reversed to your A/C axis",

but i do not know what this exactly means/translates to or how i apply it.

Thanks

[jahnj0584]

Its hard to wrap your head around on symmetric parts without a vice modeled. I know on the Haas as well, I sometimes get a -90 instead of +270 which makes you end up doing 3/4 of a turn left instead of 1/4 turn to the right. Its something we're trying to figure out.

At the end of the day, it doesn't really matter. With our trunnion, I want ALL A-axis moves to be positive so I can actually see the part (as it faces towards me).

[benjamin396SN]

So there is not really a solution then?

Because in the end, if i program the tool orientation in the correct way/orientation, the machine does mill the part correctly.

However in the case of the last 3 images when i have to mill a plane of 15°, where the A-axis has 2 possible solutions

(A-axis can move from +30° tot -120°), i cannot possibly know what fusion will choose/postprocess.

So i have to first postprocess and then go check the program which A- and C-axis moves fusion posted,

to then go back to manufacture and make sure the tool orientation is in the correct way.

So a possible solution is to adjust the postprocessor to always prefer negative A-axis moves

(part always facing towards me) so i am always sure?

(not ideal but it works i guess?)

[boopathi.sivakumar · Accepted solution]

@benjamin396SN 

you can choose the preference for a and c axis in the post so that you can easily know which way the post is going to output 

  if (true) { // note: setup your machine here => Benjamin Wijnants: false naar true
    var aAxis = createAxis({coordinate:0, table:true, axis:[1, 0, 0], range:[-120, 30], preference:0});
    var cAxis = createAxis({coordinate:2, table:true, axis:[0, 0, 1], range:[-360, 360], preference:1});

preference :0 means no preference this is default and

preference :1 means positive direction

give it a try and let me know the feedback

[DarthBane55 · Accepted solution]

I understand more what you mean now, we had that issue some time ago as well.  So we did set a preference for the A axis, now we know what to expect.  Otherwise, I believe the post will output what is closest...  meaning, if it can do for example A15C0, or A-15C180, it will output to the closest of the two, based on where the tool is at that moment.  This is from memory, and that was a couple of years ago when we ran into that, and this was the explanation I got from my post editor master guru.  In real life, it doesn't really matter that much since both will be correct, but it's better when you know what to expect which is why we forced the post to the preferred A side of our choice, like Boopathi said above.

[benjamin396SN]

@DarthBane55@boopathi.sivakumar 

Thank you for the explenation!

I will set the aAxis preference back to -1 so that the part is always facing towards me,

and i always can correctly know what the correct tool orientation must be.

Perhaps a suggestion to autodesk then is to make a function in fusion when you do 3+2 milling in manufacture,

and it happens where the possible postprocessing output is 2 solutions (like in my example A-15 C0 or A15 C180),

you can check a box if you want a preferred positive or negative A, B or C axis rotation?

Kind regards

Benjamin