please help, I can't get f360 to make working gcode, it just fails every time saying " Direction is not supported for machine configuration" How am I supposed to set it up and orient the part?
Have you selected a 4 axis machine in Setup? Make sure that the location of the 4th axis matches your machine.
yes, I have a 4 axis machine selected and even when the stock lines up with the machine it has the error, I've tried turning the part but nothing works
So there is a problem before it tries to generate the GCode. There are 3 ops in the screenshot, one with an error (a 3D Flat?) - do the other two simulate OK? Going back to this: Have you set the attachment point in the setup? (Sorry if I'm only pointing out the obvious!)
how do I make the machine face the right direction? I already tried to edit the machine and swap the axis around so the rotary axis should be along Y, it didn't do anything though
I have the Autodesk generic 4th axis machine selected, both in the Setup and in the NC program and gcode is generated ok. Note that the X32 supports only an A axis, that must run parallel to the X axis this is my machine definition, which is modified form a generic 3 axis router to add the 4th axis on the table and it will also generate gcode without problems. (still working out how to get machine simulation working....)
OK, you've chosen a machine model with 4th axis on A. By definition this is rotation around the X axis (A,B,C == X,Y,Z). There *are* ways of reconfiguring the machine, but it would be a lot easier and quicker to select one of the machines with the 4th axis on B - there are four to choose from in the screenshot in my 1st post above - check carefully which configuration matches your machine. Which controller do you have, and which post processor are you using / intending to use? Note that the OpenBuilds Post only supports an A axis (per David's post above).
Bit of an aside, but I found that this video explained it very well when was trying to get my own 3 axis machine working in simulation (from about 8:10 for the kinematics): Anyway, back to the OP, sorry.
why is it trying to go through the part? it's turning the 4th axis but also trying to go around it, the simulation works though
maybe you used 2D operations? they are not model aware, only the 3D operations can properly avoid the model
Are you using wrapped toolpaths? (I was under the impression that these only supported 2D operations, but I know quite a bit has changed in CAM, and I haven't been paying attention to the >3 axis stuff). You may need to use patches to constrain the tool path as described here:
If you post the F360 file our wonderful volunteers can take a better look for you, without needing to guess. Or at least provide all the relevant info
The "old" way of generating that sort of toolpath was to use a 2D pocket with a wrapped toolpath - I can't access 'Rotary Pocket' toolpath with the personal (free) version of F360, sorry.
I don't see why it works perfectly when simulating with machine but does something else after I post process, shouldnt the actual machine do what it does in simulation?
Are you using the same post processor for the machine setup as you are when generating the gcode? If not, you could get different behaviour. The generic Autodesk 4 axis machine defaults to their 'machinesimulation' post. You can change this to the post processor that you will be using to generate the Gcode. However! Not all post files support machine simulation: The generic Fusion360 GRBL post processor does support simulation (but not the 4th axis), but I don't believe the OpenBuilds GRBL post processor does (yet?). I would expect that you can still use it with the 'simulate' toolpath option (from right click menu) rather than the 'simulate with machine' option, but I haven't tried this.
That didn't fix it, also I am able to simulate with machine still but the gcode still doesn't do the same thing as the simulation, it has no idea how to use the 4th axis after post processing and would immediately crash into the stock because it's trying to go through it
this post shoudl generate good gcode, but DOES NOT have simulation capability at this time, like I said, I am working on it, slowly, since I am on leave and have all of my late mothers stuff to sort through....
I was just hoping to get this figured out by the 10th because I paid $200 for the 4th axis capabilities and haven't gotten any working gcode from it
Have you tried any other post processors? I can't vouch for it, because I can't test it, but came across this relating to GRBL multi axis / Fusion 360: GRBL Multi-Axis Post Processor - Autodesk Community
there are so many reasons why that can be...... from the drawing itself to the setup and the operations in it, everything matters. I have successfully implemented simulation in the post BUT found that the actual machine selected makes a huge difference to the Gcode generated. I selected the generic 4 axis mill and it gave me Gcode with crazy offsets, it would certainly crash my actual machine. So if you select the machine below it will simulate but generate invalid code for a router with 4th axis. When I select a generic router with 4th axis I get good Gcode. Right now I don't know where to look to find the differences except that the generic mill thinks it can use tool length offsets (and there is some other offset somewhere that causes Z to plunge too deep) and the generic router does not. While using tool length offsets with GRBL machine is not impossible, it is difficult and not generally done, so we need to tread carefully. So, if you have selected a machine, unselect it and let the post provide its internal machine definition, that sohudl generate good gcode. but first, fix the drawing, the paths shown BEFORE posting must be correct to have any hope of getting correct Gcode.
