Need help getting Fusion to post out correctly for 5 axis on a Haas VF3 TR mill.
When I ran this code in the mill, the thing I noticed was the B AXIS was rotating in a positive direction, when it needed to be a negative direction. (I edited that by hand to run the part.
But it seems to be aprox .250 off in what I would call the "Y" axis when A axis is sitting at 90 deg.
Hope this makes sense... 😕
I am attaching the file.
And a picture of the outcome.
Quick note on model set up.
I have placed everything exactly how it "physically is" in the mill, the same inside Fusion 360
Just looking to be pointed in the right direction...
Is it a post issue?
Is it a settings issue inside Fusion?
Good morning!
Yes trunion on the table...
I am thinking I have a post issue the more I look into this.
My issues are "B" axis is going the wrong direction.
And for some reason my Z axis is 1/4 inch too deep in the minus direction. And Y axis is 1/4 inch to deep in the minus direction.
We do 5 axis work all the time. but we are using Surfcam and have a 5 axis post for it and it is all dialed in for us.
I have purchased my own personal seat of Fusion 360 Ultimate.
And I am giving Fusion a try with 5 axis. (One would think Ultimate would have a working 5 axis post for Haas TR mills)
I have verified every offset in the Haas. It is correct.
The 3 ways that I know how to do 5 axis is...
1) Is programing the part just like you would a 3 axis job with your origin where you decide to put it.
Then having your set up guy tell you the shift amount from theoretical center of rotation of XYZ axis to the origin on the part.
The post would prompt you to enter those shift values
Let the post do all the calculations and post out the correct code.
2) Is mimic exactly what you have going on in the mill. Part location to center line of rotation. And place your model exactly like that in Fusion.
And post that out .
3) IF you are lucky enough to have a Dynamic work offset machine then you can do this all right on the shop floor.
Hello!
If your B axis is going in the wrong direction, you may need to change the machine configuration for your post process from -1 to 1 (or vice versa). You're looking a line like so:
Your other issues are probably related to trying to fix this issue by hand, so I would try fixing the machine configuration first, and then if you are still having issues, we'll go from there! 🙂
Best,
-Xander Luciano
At the time you were posting this I did find a Haas TR post and I tried that post.
It rotated in the right direction, but still too deep in z and y is still too shallow
This helped helped pointing me in the right direction for the correct rotation thank you.
The post you have shown did not work correctly but I copied the one off the Haas Trunion post off the Autodesk Cam Post page and that worked perfectly.
HERE IS THAT INFO...
if (true) {
var aAxis = createAxis({coordinate:0, table:true, axis:[-1, 0, 0], range:[-30-0.0001, 120+0.0001], preference:1});
var bAxis = createAxis({coordinate:1, table:true, axis:[0, 0, 1], range:[-999.999, 999.999]});
machineConfiguration = new MachineConfiguration(aAxis, bAxis);
BUT I still have the Z axis issue and Yaxis issue.
I had my programmer here at work post out on Surfcam the same scenario and he posted out z values .1286 different than what I am posting out.
So this seems to be a pivot point calculation that Fusion needs to be doing in the post processor.
Now our Haas Trunion center line of rotation is ( -.1286) below the platter height (Table height)
Without taking this pivot point into account it seems to be why I am having these other axis issues.
Is there anyone out there actually doing 5 axis with Fusion and uses Haas TR mills?
The first problem with the post should be solved with @xander.luciano input.
As @Steinwerks mentions your clearly missing a key aspect to the setup.
I do not have your machine. But have extensive 5 Axis experience. And what I will describe may help you.
As for the 5 Axis setup... There are typically two basic methods to manage that.
1) The first method is very basic and static in nature. That is the WCS and part program zero would be static and fixed. And typically based on the intersection of the rotary tables. What that means is you must model your fixture and part to use that as part of it's math and setup. Next the setup on the machine must precisely match that setup. The post simply looks at the WCS as the absolute origin for all math output. If the Z Axis vector that is used for the setup remains the same there is no change in output with respect to rotary motion. IF you a Setup in Fusion CAM with a Z Axis vector and then have tool paths that have a different Z axis Vector (Tool Orientation) Then the two vectors are used by the post to calculate the correct rotations for output code.
This method typically works with Table on Table method and not necessarily for Head on Head 5 Axis machines.
In the most simplistic sense you should be able to setup both Fusion and your machine to work in this mode. It's more work on the front end as you must pre determine the center of rotation and build this into your Fusion CAM setup. And insure it matches the machines actual setup.
2) The second method uses TCP that is Tool Center Point Programming. Your machine must support it. And this allows for one origin to be established in the fusion side of things. On the machine side of things your setup can be anywhere like a normal use of fixture offsets. The post still calculates the differences in vectors and the output looks pretty similar as far as rotary angles are concerned. But the output WCS does not have to be the rotary intersection. The machine handles rotary compensation in this case. There are many other advanced 5 Axis modes depending on the control manufacturer and machine builder. As seen below in this video
https://www.youtube.com/watch?v=4Nv6vgNBv8g&feature=youtu.be
I would get with your SurfCam programmer and try to mimic his exact setup in both software and then on the machine.
In the post below are a couple examples here of Fusion CAM files for 5 Axis 3+2 style programming using mehod 1) above.
https://groups.google.com/forum/#!category-topic/pocket-nc/anfiQrBirvU
I hope this helps.
🙂
Randy Kopf
http://desktopartisan.blogspot.com/
@jon.schaeffer wrote:
BUT I still have the Z axis issue and Yaxis issue.
I had my programmer here at work post out on Surfcam the same scenario and he posted out z values .1286 different than what I am posting out.
So this seems to be a pivot point calculation that Fusion needs to be doing in the post processor.
Now our Haas Trunion center line of rotation is ( -.1286) below the platter height (Table height)
Without taking this pivot point into account it seems to be why I am having these other axis issues.
To me this sounds like you need to specify the WCS .1268" below the platter face to get accurate results. Since in a table-table machine all the axes are interconnected, I would expect a small error in one axis to replicate itself in every rotation.
I don't have much multi-axis experience but that's the first thing I'd go after in this situation. .1268*2 is awful close to half of that 1/4" error.
I agree with what you're saying 100%
My experience with 5 axis is setting up Haas 5 axis TR mills and I have been doing it for almost 10 years...
I admit I only know how "WE" do it here in our shop and it is how I was trained. (Right or the wrong way I couldn't tell you) but I have always gotten great results.
So...
As it stands I changed it all up how we do it.
And started from ground zero and worked it up from there.
Turns out we have been setting our G54 Z0 wrong all this time. (Thanks to the old programmer that used to work here.)
And I was making the mistake of where to set my WCS because of the above programmer.
With your help @RandyKopf @Steinwerks @xander.luciano
BAM! Got a good part!
Thanks for the help.
Awesome news @jon.schaeffer!
Out of curiosity, how were you setting it up before? Was the post in your other CAM set to correct for the setup? Now I'm VERY curious!
Yes we had that known value inside Surfcam of -.1286
It happened behind the scenes in the software so it was easily forgotten about.
And the other thing was we probed the top of the table for our G54 Z0 We didn't use the actual pivot point of rotation
So this required the shift.
So this still brings me one question maybe you can answer.
Going with this way of 5 axis you all have me going. What do you do in this scenario?
Second op 5 axis part.
In a 4" Vise.
WCS is
X0 = left side of finished square part (against mill stop)
Y0 = Hard Jaw
Z0 = Top of Parallel's
Now the vise is always going to be placed in various spots on the table at every set up.
And so will the vise stop.
IF this was a production job do you need to Physically move the model to how the set up tech did at the time of set/up every time you run the job?
Or is there another way?
@jon.schaeffer You should be able to put it anywhere if your second setup is on the table since there would be no rotary to contend with. If you can get it real close you could have the machine probe the part and/or workholding to set your offsets and keep things real tight. At least I hope you've got a probe on that machine!
Glad you got things figured to get a part working.
Your latest question is all based on whats supported on the control.
In my shop at http://www.rmbproducts.com/ I have a CMS Router that fully supports setups exactly as @Steinwerks suggests.
BTW SpaceX has 15+ CMS Routers in a much larger scale used in this exact same scenario to build many of their rocket parts.
We are able to use standard Fanuc Fixture offsets G54-G59 or G54.1-G54.99
We have multiple parts in different locations all managed through a unique fixture offset.
We probe only when we have to. Our parts are rotation molded plastics and we do 5 Axis simultaneous trimming. If the parts have variance AND we have to control relative locations we probe. ALL Compensation is managed through TCP via G43.3 on our machine in conjunction with the previously mentioned Fixture Offsets.
Hopefully your situation is similar. If not you may be forced to use the static method I previously mentioned.
You part or multiple parts can still be anywhere in the working travel envelope. The trick here is the WCS Origin must be modeled with respect to your center of rotation. And the shift of where the part and its origin must also be determined. Then on the machine each part must be setup with the same shift replicating the offset.
My hobby 5 Axis machine works this way at home. My machine supports 3 Axis Fixture offsets but it does not do 5 Axis compensation other than with the G54 Fixture Offset.
I hope this helps!
🙂
Randy Kopf
http://desktopartisan.blogspot.com/
I am having issues with the Next Gen Post for a Haas TRT 100. I can't seem to get the sign direction to post correctly for the A Axis. When I switch from a (1, 0, 0) to a (-1, 0, 0) on the A axis configuration it posts a sign change for my C axis.