Faceted finish on curved surface

[frankTN5LB]

Hey everyone, 

My name is Frank, I'm new to fusion and fairly new at 3d profiling. I got this odd job to make a simple mold and needed to 3d mill the shape. Could anyone tell me why I'm getting this "faceted" finish instead of a smooth curve? I noticed in the simulation the points on the toolpath line up with where the facets appear. Just not exactly sure why they're there. I know in Mastercam there was an option to randomize points, but I didn't see anything like that in fusion. Could anyone tell me why this happened and how I can fix this issue?

Thanks

[seth.madore]

Any chance you could share the file?

File > Export > Save to local folder, return to thread and attach the .f3d file in your reply.

Was this a native Fusion file, or was it modeled in another CAD package?

[frankTN5LB]

Hey Seth,

Thanks for taking the time to respond, unfortunately I'm way from the shop and don't have a way to share the file at the moment. I can upload tomorrow. Thus was an imported model, I drew it in solidworks and imported as a step file. 

Frank

[frankTN5LB]

Things have been busy, running more and more parts that require 3d surfacing and this is still a problem I'm running into. I'm attaching the original file from when this thread was started as well as the file for the most recent part. I make my models in solid works and import them into fusion 360 for cam. I have tried directly opening the solid works file in fusion, as well as saving it as a step file and opening it, all with the same results. I have also tried drawing the models in fusion, same result. I am now trying to use the ramp toolpath to add a 1 degree taper onto a curved surface and am getting the same faceted finish. I talked to Autodesk support about the issue and they walked me through adjusting the smoothing and tolerance settings but that didn't change the resulting finish. Support ended with an idea that it might have to do with the machine look ahead? Also I did notice that the lines of the facets correspond to the points on the toolpath. One last note, the machine is a 2012 Haas VF-1. 

[leo.castellon]

In my opinion, being a long time Mastercam for Solidworks user, when I had this problem, it usually happens with splines and ellipses. Looking at your mold program, I would tighten up the tolerance, like to .0002. That should eliminate a lot of the faceting. If you run the simulation on the scallop program using .002 versus .0002 for tolerance and smoothing, you can kind of see the effect if you zoom in closely.

LeoC

[frankTN5LB]

I can give that a try on the parts I'm running now. Before I start though, I understand that splines and ellipses can cause some problems and I know that a lot of the geometry of the mold ends up being splines when the cam breaks it down. But does that explain the the issue with the ramp toolpath on the conical surface? Maybe I'm just not wrapping my head around the geometry but in my mind that should be a bunch of G03 arc movements at incrementally increasing depths and radii. 

Frank

[HughesTooling · Accepted solution]

Arc fitting (Smoothing) will not work if more than 2 axis are moving. It tends not to work well with scallop unless it's on a flat face. It will not work at all with ramp on a curved face, with your sample part you'd get just as good a toolpath using 3d contour set to both ways and as it parallel to the XY axis you'll get good arc fitting. The ramp toolpath is already cutting both ways so no advantage on this part.

You can see here the 3d contour toolpath is only 20kb, see file attached.

[frankTN5LB]

You are right on. I actually switched my toolpath to use contour instead of ramp and the entire problem went away. I played with a few other toolpaths to see what was actually happening and why contour worked. I guessed that 3d motion and smoothing weren't playing well. In this instance the constant z allowed the fusion to post arcs instead of lines and the entire problem disappeared. Choosing the right toolpath is more important than I thought. I was about to post my findings here and saw your post so it is nice to have my hypothesis confirmed! Thanks everyone for your help and input!

Frank