-
F360 - spiral toolpath on 4th axis?
Hi all
Whilst Fusion 360 has continually impressed me with the quality of its toolpaths and just how easy they are to create, its support for 4th axis continuous machining has left me either confused, disappointed or both. As far as I've been able to find, it will let us wrap 2D toolpaths around a cylinder centred on the axis but that's about it.
Here's a a couple of typical ops I'd like to perform but that I can't work out how to set up and I'm hoping someone out there is going to reply with "Idiot: it's easy, just select strategy X and enable option Y." :)
1. "surface" a cylindrical face which is not concentric with the 4th axis. For example, if I want a piece of flat bar stock which has a slight belly on one side. At the moment I am stuck with indexing the 4th to set that face upward then doing a 3D path. I'd kinda like to get that 4th axis live for this operation to keep the tool tip normal to the model surface. Easy with the swarf op if you have a 5A machine, but nothing for "just 4th"?
2. Spiral cut around a form. Using the tip of the tool, again normal to the model surface, I'd like to be able to start at one end of the workpiece and spiral up to the other, cutting the outside of a smooth model. Sort of like a contour strategy wrapped around the part.
Am I thinking about all this the wrong way? Or have I missed something simple and obvious for doing it? Or have Autodesk just not implemented 4A support?
Thanks in anticipation of any help!
-
Re: F360 - spiral toolpath on 4th axis?
If you can't get Fusion to do it, try DeskProto,
-
Re: F360 - spiral toolpath on 4th axis?
How savvy are you with programming, dharmic? Another question is how flexible/limited is this F360 PP modification?
I could be wrong but I think F360 uses C (jscript) to output it gcode. Meaning you could unwrap a surface (sure F360 has this command?), create a 3D toolpath on that, and get that to wrap back onto a OD 'via code' during post processing.
You'd lose the simulation and all, a whole lot of work & testing too, but your gcode would output what you want.
Not exactly The Solution for a one-off part but if you mainly 4ax then I'd look at PP mods (if allowed, obviously)
-
Re: F360 - spiral toolpath on 4th axis?
Hey Mecanix. I'm savvy enough to have cobbled together my own post for Mach3 with 4th and some probing routines. A while ago so it's fuzzy now but I seem to remember it's in javascript.
I think it'd make life harder - trying to map where on the flat surface various things belong on a non-cylindrical wrap seems like a recipe for pain.
After a half a day exploring yesterday, I think there might be some promise in the Flow strategy which I'll try today. Otherwise I foresee a bunch of sketch offsets turning into a toolpath just to get this job done :D
-
Re: F360 - spiral toolpath on 4th axis?
I've done the 3D surface contour (re)warp onto 4ax in C# (nxopen), works okay other than the overhead processing time. Method/vectors/rad could easily be converted to fit in Jscript too I'm sure so let me know if you see an entry in the PP where that could done. I'll see if I can give you a hand with that.
Wish I knew F360... all mysteries to me :/
-
Re: F360 - spiral toolpath on 4th axis?
Thanks, Mecanix - that's a very kind offer.
But, to my mind, the post should be nothing more than a translator from CL to machine specific code. Putting smarts in there where it can't be tested easily sounds like a trap for future dharmic who has forgotten whatever cleverness I might do today.
I'll keep playing with a combination of adaptive clearing and the Flow tool. The big trick here is in trying to minimise torque on my 4th which is just a gutless little stepper on a 3:1 or similar belt.
-
Re: F360 - spiral toolpath on 4th axis?
Quote:
Originally Posted by
dharmic
But, to my mind, the post should be nothing more than a translator from CL to machine specific code.
I rarely heard of any package translating correctly out-of-the-box, pretty standard unfortunately. Whether that's intentional, who knows, although many claims it is i.e. they wouldn't sell as many 5ax licenses and/or custom PP if we could do full sync 3D on 4ax with only a few clicks :/
Obviously unhelpful and depressing^. Sure someone with hands-on exp will chime in soon.
-
Re: F360 - spiral toolpath on 4th axis?
I thought that you could already do wrapping with Fusion 360?
But wrapping is far from ideal in many situations.
And as far as I know, Fusion still does not have full rotary axis support. Which is strange, because it can do 5 axis.
https://forums.autodesk.com/t5/fusio.../idi-p/8678700
-
Re: F360 - spiral toolpath on 4th axis?
Gerry, "wrap" is available only for a couple of 2d strategies and it only allows wrapping onto a convex cylindrical form. As soon as the form is non-cylindrical (or, as on one of my ops, on a concave cylindrical face) you're hosed.
Yeah, the more I look into it the more I see other people b1tching about the lack of support and at least a year's worth of teasing from Autodesk hinting that it's "coming soon".
With Flow I can be careful about my U/V selection and it kind of works but it's a finishing-only strategy. I'm probably going to end up building an envelope with a decagonal(?, ten sides anyway) profile and do a bunch of adaptive (trochoidal / hsm) hogging paths around the part to restrict how far off-axis any of them get - just trying to keep the torque under control. Then finish off with the Flow strategy. Or something.
Side note. Despite it costing a billionty times as much, I now remember having the same issues with Creo's CAM package - 3 axis no worries, 5 axis no worries, 4 axis continuous was basically "Yeah draw some lines and tell the tool tip to follow them". Some real deja vu going on right here. Major difference is that F360 is infinitely cheaper and just as infinitely easier / saner to drive in the stuff it does support.
-
Re: F360 - spiral toolpath on 4th axis?
Quote:
Originally Posted by
dharmic
Gerry, "wrap" is available only for a couple of 2d strategies and it only allows wrapping onto a convex cylindrical form. As soon as the form is non-cylindrical (or, as on one of my ops, on a concave cylindrical face) you're hosed.
Yeah, the more I look into it the more I see other people b1tching about the lack of support and at least a year's worth of teasing from Autodesk hinting that it's "coming soon"..
I think you need to understand more how fusion works... which is on SOLIDs and not MESHes. So as long as you have a face or edge on a cylinder you can machine it. Both the examples you gave can be easily achieved.
But in the case of wrapping, you flatten a mesh. Fusion has no way of machining meshes. If you really wanted to wrap on fusion you would need to convert the mesh to nurbs and machine the model as a whole.
-
Re: F360 - spiral toolpath on 4th axis?
Quote:
Originally Posted by
ger21
Fusion 360 has full 4th indexing (positional) support, 3 + 2 5th axis (positional) support and 5th axis simultaneous support.
Re wrapping, it doesn't make sense to wrap on fusion as per my other post, fusion is not a mesh based program.
-
Re: F360 - spiral toolpath on 4th axis?
Thanks, FastFarmer. I certainly do need to understand more how fusion works... which was kinda the point of this thread.
BTW: Fusion does offer a wrap function without needing to even think about meshes, allowing 2d ops on convex cylindrical wraps. What it doesn't allow is for that wrap to happen on non-cylindrical forms. No help to me.
I don't want to wrap. I want to do continual 4th on the face(s) of a non-cylindrical form.
Not offered by Fusion currently, being teased and hinted at coming though.
In the meantime, my strategy of hogging using 3D Adaptives around 8 indexed containment areas then finishing off with carefully constrained flow ops is simulating well. Will try and throw it on the machine tomorrow and see how she chooches.
-
Re: F360 - spiral toolpath on 4th axis?
Quote:
Originally Posted by
dharmic
I don't want to wrap. I want to do continual 4th on the face(s) of a non-cylindrical form.
In the meantime, my strategy of hogging using 3D Adaptives around 8 indexed containment areas then finishing off with carefully constrained flow ops is simulating well. Will try and throw it on the machine tomorrow and see how she chooches.
This is the correct method. When you post and run on the machine, it will be continuous and in 3D. Just make sure for each index your setup origin is in the correct position/direction. If you need, use a sketch to do this ;-)
Your 2D "wrap" works because its a line or face on a plane... it's not a 3D wrap, simply a 2d flatten.
-
Re: F360 - spiral toolpath on 4th axis?
lol... now this subject is on my mind dharmic!
Wanted to also add, although wrapping may sound convenient it's often not. The main issue is in 3d wrapping an axis (typically Y) is substituted with A. In other words the machine still operates in a 3 axis manner. So for example if you wanted to machine a flat face on your cylinder, it's pretty well impossible. Positional indexing allows this and complex shapes. It uses all 4 axis, so although it requires more CAM work, it's absolutely the way to go.
Hopefully this helps. Cheers
-
Re: F360 - spiral toolpath on 4th axis?
That's not "continuous" 4th axis machining though. It's a lot of 3a operations done on an indexer.
Meh, it'll get me through until Autodesk release real 4A support on 360. :)
Cheers
-
Re: F360 - spiral toolpath on 4th axis?
And yeah, I've played with the wrapping stuff before - it can be enormously helpful in some cases, keeping torque loads on the 4th nice and low by primarily working in close to the axis. And a lot of 4A useful operations can be done with cylinder wrapping.
Just not what I want in a lot of cases and I made the flawed assumption that, if they had continuous 5A support, 4A would be kind of a gimme in the bundle.
We live and we learn :D
-
Re: F360 - spiral toolpath on 4th axis?
Well... if your processor supports it, then there's no reason why you cant set up a dummy 5 axis and as long as you program your tool position square to the face, form a continuous 4th axis machine wrap.
Might even have to try that myself... lol
-
Re: F360 - spiral toolpath on 4th axis?
Quote:
Originally Posted by
dharmic
We live and we learn :D
+1
Tempted to dwload and give that F360 a first-time tryout. When I'm out of project I'll do that... let's hack that post for 3D cylindrical projection. Surprised nobody did it yet... math/formulas is all public domain stuff
-
Re: F360 - spiral toolpath on 4th axis?
Quote:
Originally Posted by
Mecanix
+1
Tempted to dwload and give that F360 a first-time tryout. When I'm out of project I'll do that... let's hack that post for 3D cylindrical projection. Surprised nobody did it yet... math/formulas is all public domain stuff
Dislike 360 Design but 360 CAM is unbeatable... so get on board!!
It hasn't been done because as I have tried to explain, Fusion works on faces, not mesh models. Even when 5 axis simultaneous machining. you are still machining the surface of a face of a SOLID... not the model or mesh, and you (the human programmer) must consider things such as tool position, angle to the surface, and collision... for that particular face..
But I wouldn't want to see it go any further towards AI in 360, and legally I think it would move liability to the software which would be a bad thing. If I want to wrap a model, I use Vetric which does a great job at it.
-
Re: F360 - spiral toolpath on 4th axis?
Familiar with the cad topological data i.e. curves, vertices or faces (brep). Mesh (polygons) not so much though :/
If the software allows for unwrapping of a surface then you simply need to machine this surface normal to Z with a zig zag pattern and wrap back those points it outputs to the PP via code. To my 'limited' knowledge F360's PP is C powered so it can't be that complicated to code. Bet its been done by a few already, they are just not telling us ;-)
-
Re: F360 - spiral toolpath on 4th axis?
Quote:
Originally Posted by
Mecanix
If the software allows for unwrapping of a surface then you simply need to machine this surface normal to Z with a zig zag pattern and wrap back those points it outputs to the PP via code. To my 'limited' knowledge F360's PP is C powered so it can't be that complicated to code. Bet its been done by a few already, they are just not telling us ;-)
You're not understanding the difference between a mesh and a solid. Fusion cannot machine meshes... so they are not telling you because it's impossible.
You can convert your mesh to a solid that fusion can understands, but then you cant unwrap a solid... it's a solid. You can unwrap a mesh because it's a shell... it's simply a surface with no volume. If you unwrapped your mesh and then converted it to a solid so fusion could machine it, then it would be a solid again. So, there is no way to get around this issue.
In the same way, Vectric can unwrap a mesh but it cannot machine a solid. Thus, Vectric can wrap on the 4th axis, but it cannot index on the 4th axis.
These limitations are because of the models used and the associated advantages/disadvantages of both.
-
Re: F360 - spiral toolpath on 4th axis?
... but you CAN extrude a new surface (thin feature) 0 distance from a face, and use that to define a tooling boundary. If it's on a cylindrical face you can then click the checkbox on a 2d strategy to wrap it.
-
Re: F360 - spiral toolpath on 4th axis?
Pardon my total ignorance of F360, @FastFarmer. Didn't know you could model meshed surfaces with F360 at the first place. Kinda strange though as I just can't understand how they manage to fit those polygons within that parasolid kernel, which I think is what Autodesk's Fusion & Inventor are licensed with (?).
So much to discover, so much to learn, so little time :/
edit: I've just gone google-crazy and it appears as if you can work meshes in F360. Unbelievable!!! And no bloody wonder parts can't be machined LOL holly snap man....
https://knowledge.autodesk.com/searc...usion-360.html
-
Re: F360 - spiral toolpath on 4th axis?
You can model/work with meshes in fusion, but just not apply meaningful toolpaths to it. You need to convert it to a solid first; which you can also do in Fusion 360.
-
Re: F360 - spiral toolpath on 4th axis?
Quote:
Originally Posted by
FastFarmer
....which you can also do in Fusion 360.
No Thank You Very Much! Can't do hybrid CAD, or whatever its called backend-wise. Never had training for that lol
Need standard CAD nurbs/splines to work out surfaces/edges, prefer control over the points where the stuff sews together. Machinable, at least ;-)
Call me old-school if you want, don't care...
Thanks for the heads up, didn't know. Saved me quite a few hours in fact
ps: a CAD that really caught my attention recently is Freecad. Picked up on this bad boy as a python learning platform during my free-time and I just can't get enough of it. Bet it could cut dharmic's part even with 8axis 'simultaneously' with little prog effort...
-
1 Attachment(s)
Re: F360 - spiral toolpath on 4th axis?
Thought I'd add my voice to the mix in the event that Autodesk is listening. Attached is a chair part I'd like to mill on a 4th axis. I believe the dowel hole and the saddle joint on opposite ends could be easily done via indexing. But the real challenge is the second operation mounting the part between centers and final milling the surface in one pass. And much like 5 axis continuous , tilting the ball end bit at something other than surface normal to optimize cutting geometry for a clean cut. Fusion 360 should be able to address this (hopefully). Maybe not now, but soon? I know there are ways to do this now via indexing, but it would require some additional geometry and there will always be visible transitions, one side to the other. If anyone out there knows of other options for doing this, I'd love to hear them.
Marv
-
Re: F360 - spiral toolpath on 4th axis?
Quote:
Originally Posted by
MARV
Fusion 360 should be able to address this (hopefully). Maybe not now, but soon?
Nope.. it's impossible. In this case I would think fusion will not recognize this shape as a cylinder and there is no face to wrap.
I will have a play with this in the next day or two, but you need to appreciate fusion is an engineering based tool, not an art based one.
There are a couple of mesh based cams that will wrap this no issue.
-
Re: F360 - spiral toolpath on 4th axis?
Seems to me it would be easier to just work out the maths and write your own g-code.
(Meant quite seriously.)
Cheers
Roger
-
Re: F360 - spiral toolpath on 4th axis?
The chair part in question was done in Rhino. Not an engineering program per se, but quite capable as a solids modeler. And I've used Rhino Cam in the past to do similar surfaces. Retired now, I don't have the $$$ for Rhino Cam, so Fusion is probably going to be the tool of choice.
I'm a ways off to a 4th axis for my machine, so I hope that further development in Fusion will come by the time I'm set up.
Marv
-
Re: F360 - spiral toolpath on 4th axis?
My suggestion would be if you want to machine these type of objects on a forth axis, then go get yourself a copy of vcarve.
-
Re: F360 - spiral toolpath on 4th axis?
For wooden parts like that, get Deskproto, which can do continuous rotary. But I'd still do the joinery as separate toolpaths with indexing.
People have been asking for continuous rotary in Fusion 360 for a long time. I don't see it coming any time soon.
-
Re: F360 - spiral toolpath on 4th axis?
Quote:
Originally Posted by
FastFarmer
Nope.. it's impossible. In this case I would think fusion will not recognize this shape as a cylinder and there is no face to wrap.
I will have a play with this in the next day or two, but you need to appreciate fusion is an engineering based tool, not an art based one.
There are a couple of mesh based cams that will wrap this no issue.
I had a play with the multi axis flow tool path last night. As suspected it wont recognize shapes like this as a cylinder, so forget it.
I guess it's worth also pointing out that this limitation doesn't just apply to 4th axis, but also 5 axis. Again this limitation is due to the model type being a solid. So if you thought wow, I can simultaneous machine my model on 5 axis with fusion 360.. you can forget that too.
Interesting discussion. It shows that although fusion has been smart in cornering the market with its free software, there is still a future for hobby type mesh based modelling cam with companies like Vectric and Deskproto.
-
1 Attachment(s)
Re: F360 - spiral toolpath on 4th axis?
Quote:
Originally Posted by
dharmic
Hi all
Whilst Fusion 360 has continually impressed me with the quality of its toolpaths and just how easy they are to create, its support for 4th axis continuous machining has left me either confused, disappointed or both. As far as I've been able to find, it will let us wrap 2D toolpaths around a cylinder centred on the axis but that's about it.
Here's a a couple of typical ops I'd like to perform but that I can't work out how to set up and I'm hoping someone out there is going to reply with "Idiot: it's easy, just select strategy X and enable option Y." :)
1. "surface" a cylindrical face which is not concentric with the 4th axis. For example, if I want a piece of flat bar stock which has a slight belly on one side. At the moment I am stuck with indexing the 4th to set that face upward then doing a 3D path. I'd kinda like to get that 4th axis live for this operation to keep the tool tip normal to the model surface. Easy with the swarf op if you have a 5A machine, but nothing for "just 4th"?
2. Spiral cut around a form. Using the tip of the tool, again normal to the model surface, I'd like to be able to start at one end of the workpiece and spiral up to the other, cutting the outside of a smooth model. Sort of like a contour strategy wrapped around the part.
Am I thinking about all this the wrong way? Or have I missed something simple and obvious for doing it? Or have Autodesk just not implemented 4A support?
Thanks in anticipation of any help!
Use the rotary toolpath.
https://www.cnczone.com/forums/attac...d=441562&stc=1
-
Re: F360 - spiral toolpath on 4th axis?
-
Re: F360 - spiral toolpath on 4th axis?
New toys! Thanks schneik, exactly what I was looking for back in September. Grump grump "where is this rotary thing" this morning, did an update et voila - simple and exactly what I was after. Whether the latest version took it out of experimental mode or it was how I picked up the freebie manufacturing extension I don't know, will find out in June I guess :D
-
Re: F360 - spiral toolpath on 4th axis?
Quote:
Originally Posted by
dharmic
New toys! Thanks schneik, exactly what I was looking for back in September. Grump grump "where is this rotary thing" this morning, did an update et voila - simple and exactly what I was after. Whether the latest version took it out of experimental mode or it was how I picked up the freebie manufacturing extension I don't know, will find out in June I guess :D
Yeah, thanks schneik. I've been playing with this for a while now, and although I've had to do quite a bit of fiddling with my post to stop unwinds etc, the rotary tool path on both horizontal and vertical works amazingly well.
A remaining issue I have is where a following rotary toolpath from another rotary toolpath continues from the axis position of the old one, where it would be better it reset to A0 again. Not sure how I can fix that?
-
Re: F360 - spiral toolpath on 4th axis?
There is a guide to Coordinate Spaces at the Artsoft web site which discusses this for Mach3:
Coordinate Spaces - a Guide
You can reset the A axis value in Mach3 in several ways, not all of which are useful. The following code block works on my machine
#99803=0
G28.1 A0
#99803=0
No guarantees of course.
Cheers
Roger
-
Re: F360 - spiral toolpath on 4th axis?
Hi Roger,
I use Linuxcnc, the post code in fusion for bypassing unwind is -
writeBlock(
gMotionModal.format(0),
conditional(abcFormat.format(abc.x) == 0, "G10 L20 P#5220 A[#5423 MOD 360]")
);
With this whenever you issue a A0 comand, it will insert this code before A0 and always return it to the 0 within 1 revolution (See code below)
But that's not the issue I am talking about. What I am saying is if you program one rotary path following another the code will look something like -
N74405 Z15.403 A-26305.368
N74410 Z15.401 A-26352.635
--
--
N74465 G49
(ROTARY1 2)
N74470 G53 G0 Z0.
N74475 G10 L20 P#5220 A[#5423 MOD 360]
N74480 A0.
N74485 G0 X0.098 Y0.
N74490 G43 Z31. H31
N74495 G0 A-26370.
N74500 Z5.371
As you can see the last rotary was A-26352.635 but when you start a new tool path you issue an A0 and then start from where you left off A-26370. This means you goto A0 but then must wind back.
You could remove the A0, but in the post processor this has effects on other toolpaths such as adaptive clearing that issue A0 for positioning. What needs to happen is the rotary tool path needs to start from A0 again, not the old position.
If you separate the posts then it starts from 0, but not really ideal.My work around at the moment is simply to delete A0 from g-code.
Cheers
-
Re: F360 - spiral toolpath on 4th axis?
Hey Roger
Thanks for posting that link (and the original doc). Reading through it now when I should be doing something else because ~finally~ it's all starting to come together in my head instead of being a bunch of disconnected "oh this does this, that does that" snippets. Great stuff!
Anyone else: don't forget to flick through to the last page and get the last version instead of the one on the first page.
FastFarmer just a thought: if you're resetting the machine workspace you're going to have to reset the workspace in the toolpath as well. Perhaps in the post for each of the operations? Or as a manually inserted block between them?
-
Re: F360 - spiral toolpath on 4th axis?
@farmer: Ah, I don't know LinuxCNC, so can't help.
@dharmic: I wrote that document initially to get it sorted out in MY head!
Cheers
Roger