CNC Foam Lathe

  For the last few moths I have been building a 4 axis cnc foam cutter for building wings and other uses. About a
month ago someone from the RC flying club I am with asked me if I could use the foam cutter for building round
fuselage foam plugs so he could make fiberglass molds of the plane parts . The more I thought about the
problem the more I came to the conclusion that a lathe was a more appropriate solution. Keep in mind these
pieces would be rather large not just an inch or two in diameter. They would also be complex in shape. integrated
canopy tear drop and eccentric cones type of things. This means that I would also need to have spindle coordinated
cutting and all that with a heated tip as the cutting tool.

  Unlike my currently operational foam cutter I need to use LinuxCNC for the more advanced capabilities that
GRBL simply doesn't have like closed loop spindle control. I will be using linear rails and 3D printed parts with
aluminum extrusions for the frame.

Stay tuned this could be interesting

  So my first obstacle doesn't even have to do with the machine, it has to do with my cad/cam program.
I had never needed to use more then 3 axis of movement before so it wasn't until today that I realized
fusion 360 won't allow synchronized turning without paying extra for the add-on. Keep it up Autodesk
I am really close to fully abandoning Windows and fusion 360 altogether. the Question now though
is do I just treat the models as 3D and add the 4th axis manually or can I repurpose the third axis as
a rotary axis? I guess there's always DevCAD.  More research is needed.
 

Interesting project.

Without knowing what your machine or the cutting tool (wire) will look like, would it be possible to use polar interpolation and configure the machine & F360 like a C-axis lathe?

Another thought...I wonder if using LCNC's external offsets would be a way to get around the F360 restrictions.  Andy Pugh and some others have posted some pretty interesting non-cylindrical turning videos... maybe a low-speed foam lathe would be a great platform for external offsets.

Hi spumco. I will look that over. In the case of my machine I intend to use a soldering iron for the
cutting tip.
www.amazon.com/Walnut-Hollow-Versa-Temp-...id=1721035279&sr=8-6

  So far I am thinking I may set the lathe up more like a 2 axis with the spindle as a rotary indexer to escape
the multiple axis problem. That does require a rethink for setting up the part model of instead of a cylindrical part
to a 3D flat part and cut the model out in multiple longitudinal slices.

  in any case the G-code will require splicing to bring it all together.

I think F360 could be coaxed in to spitting out all the longitudinal moves in one go.

• Draw an axis down the rotational axis of your part
• Create plane at angle, using the axis line
  • Project the cross-section of the part (body intersect) on to the plane
• Create pencil toolpath along the 'top half' of the projected path
• Repeat the plane & projection as many times around the same axis as you need resolution.
  • Would depend on the resolution of the burned/melted area your soldering iron tip creates - which likely varies depending on temp and 'feed-rate.' Might need 36 planes, might need way more.
• Repeat toolpath for each rotational angle plane, using tool orientation to adjust the "A" axis (in your case it'd be the C-axis)

Posting the gcode should result in a single file with a series of longitudinal moves (Z) with varying X values along the part profile, repeated for each A-axis orientation.

I wonder how deep you could dig with each pass? I was imagining something like a loop-shaped hot wire knife scooping off great big ribbons of foam like Godzilla's own apple peeler.

  So many good things to respond to.spumco.
While I was at work and changing out runs on the shops 3D printers I had a few chances to dig through
youtube and find some interesting examples. This one company builds a 7 axis foam cutter. Essentially
it's a 4 axis like the usual but with a turn / indexer table attachment and also a lathe attachment. It did
give me some ideas . Maybe I was over thinking the problem I may just need to build a rotary indexer
table for my already operational foam cutter plus I still have an extra axis available on the current
Arduino / ramps 1.4 board. and if I need it to operate as a lathe I just need to add some uprights.
The hard part will still be figuring out exactly how to handle the CAM end . And you may be right  the
answer could very well be slicing the cylidricle model up like a pie and just making cuts like it was a bunch
of 2D splines. Somewhere between 36 to 72 should give good enough resolution and just have the indexer
turn in 5 to 10 degree increments between cuts.

  As far as the thick hot wire for scooping out large chunks of foam at a given time, a bent 12 awg wire
only needs around 2 volts but almost 20 amps and ironically the video showcases the thick wire cutting a
coarse thread in a foam pillar .at 3:25

  Obviously the video in not mine . Some may find the music annoying  but just turn down the audio.

 

That thread cutting vid was exactly what I'd cobbled up in my mind, except I'd imagined an X-axis moving as well.

Now take that mechanism, use a less "v-shaped" wire, and add LCNC's external offsets in X to the rotary toolpath.

With some expirimentation you could create some very complicated shapes.