I feel you. Replace “little one” with “neighbor” and “basement” with “garage” and that’s where I am.
Anyway, I just wanted to highlight that the reason I mentioned the online version of @fenrus’s tool is precisely because it’s a major shortcut to avoid having to understand and deploy the other goodness he put up on github: it literally takes a few seconds to upload an STL, save the generated heightmap, and import that in CC. I remember doing exactly that in this little project of mine, picking up an STL from thingiverse.
I wish @robgrz would integrate his (or similar) code in CC Pro as an “import STL” feature, it’s a killer feature in terms of productivity.
I know I should turn the STL-to-gcode tool as a javascript app as well… it’s a bit more complicated than just writing out PNGs… (but not all THAT much… just needs time, and I am wrapping up a house move which hasn’t allowed me to have time to play lately… on the good side, the basement is a great space for my XL )
I just wanted to report that using Carbide Pro was mostly successful. I say mostly, because I had a couple of issues.
The first was that one of the lakes in the view appeared as a void in the carving. I was able to solve this by using paint and lightening up the grey in that area.
The second, is that the mountain top carves flat. I considered it was also a gradient problem. However it shows correctly in Carbide Create. My retract height is greater than my stock surface height.
Once I get this carving completed, I am totally gonna try Fenrus’s tool on one of the STLs I downloaded.
I need to order some bits. RIght now I am using my 201, .25" that came with the machine. I tried the 60 degree V bit, however, it failed to compensate for the width of the actual point. so when it was running the finish pass it consistently stepped up too soon and stepped down late, leaving the rough edges.
I totally understand the programming issue. I don’t know javascript, but I’ve done enough coding to know that simple is hardly ever a reality. No matter how simple the idea is.
also for 3D carving, a tapered ball nose bit (like https://smile.amazon.com/gp/product/B015C61XX4) are the way to go.
(unfortunately I managed to drive mine vertically deep into the bitsetter top and snap it, so Amazon will bring me a new one soon)
it’s still early, and a few hours of coding to go before I’m going to actually cut some wood with it… but the basics work and the gcode looks ok for the functionality that is implemented.
(performance is… not great yet)
the PNG intermediate loses you resolution/detail to be honest. For many cases that does not matter, but I’ve managed to carve with so much detail that I could see my own house on a 5x5" cut.
If your tool doesn’t compensate, then how are you able to get such detail with a tapered ball end? I switched to the flat end for a finish pass, because my V bit skipped entire sections of the cut when the software failed to recognize the difference. I had to choose between an unfinished carving and a carving that required a little extra sanding.
the software supports the ball geometry, but not the tapered part.
as long as you don’t have steep walls it doesn’t matter
I am wrapping up some other changes into the online version, and then will probably just add support for the tapered part; it’s not complicated or hard, just a tiny bit of trig math.
(I could also add V bit support but… not sure one would do 3D carving with that normally)
That would be cool. I can see that. I only used the V Bit because I only had it and the flat. I have been doing my best to not just buy all the bits that I see and stick to the ones I need as I go. The ball ends that you recommended will be here tomorrow.
if you’re looking for some other no-regret bits: https://smile.amazon.com/gp/product/B073TXC45Z are great for 2D work. downcut (so less sanding) and able to do pretty good detail while being sturdy enough that you can use a 20 ipm or higher feedrate (smaller bits get so fragile that you have to go real slow)
)
