Some new, bigger Ideas, just rough outlines, to be dicussed

[uninteressant2026]

Your pace of development is simply incredible, so here are a few more ideas—just rough outlines, without having thought through the details. So more like sketches.

1. Support structures. For 3D printing, you occasionally need support structures. These are generated more or less well by the slicer, but in my opinion, they’re usually poor.
   I tend to tilt objects at a 45-degree angle so that such structures aren’t necessary. However, this means you have to prop up the object because it’s standing on just one edge and would otherwise fall over during printing.
   In such cases, I like to place triangles on one side for support. The adjacent side serves as support, with very small connections to the actual object, someting like this.

Perhaps theres a way to add these as configurable objects.

2. Programming Objects link in Tinkercad
   I think theres not much more to say. To describe it here in detail is way to much.
   Just an Idea, because the possibilities in Tinkercad are limited by the small number of selectable objects.

Gruß
Ingo

[henmedia]

@uninteressant2026 thanks, Ingo, both are good to think about.

1. Support ribs: I like it, a tilted part standing on one edge is exactly the case where the slicer's supports are clumsy. I can picture it as a shape that you place against a sloping face of the selected part: a thin triangle from the bed up to that face, with a small gap and a few tiny contact points that break off. To get it right: what has worked for you so far? How wide is the gap, how big and how far apart are the contact points, and how thick is the rib? Until then the Wedge does it by hand, turned so its slope lies against the face, with a few small cylinders as the contacts.

2. Programmable objects like Tinkercad's Codeblocks: that would be a completely new field. It would have to work for every shape, every tool and every setting, and since every change would have to run the whole program again and rebuild every part, it would almost be a product of its own. Even the simpler way, where the result becomes normal shapes in the design, is a huge project in itself, far more development work than I can take on at the moment. That said, if someone wants to take this on and send it as a pull request, I'm happy to look at it, just don't underestimate how big it is. What comes closest in layerling today: numbers in every field can be formulas, your own shapes can be saved and inserted again, and the MCP interface lets a script or an AI build parts step by step.

[uninteressant2026]

ok let's forget step two so far and let's stick to the support.

[henmedia]

@uninteressant2026 good, let's stick to the supports. I'd like to try to build them the way you want them, so please describe as exactly as you can how they should look. You have surely found out by now what works and what doesn't, and I'd like to use that. For example: how thick the rib is, how wide the gap to the part is, how many bridges there are, how big they are and where they sit (only at the two ends as in your picture, or also along the slope), what makes them break off cleanly, whether the rib is solid or has openings, and for which angles you use it. A small sketch with sizes would help a lot. Or, as well as that, attach a design where you built such a support as a .lyl file (zipped, as GitHub does not take .lyl files directly), and I'll take the sizes from it.

[uninteressant2026]

I’ll try to put together a few parameters—or rather, define a few things.
We’re not necessarily talking about a triangular shape, but it’s a good starting point.
The triangular shape allows us to position objects at any angle on the print bed and support them with the triangle.
In addition, though, I can also imagine a shape that supports horizontal overhangs, for example. I’m thinking of something like a column that widens toward the top.
Maybe someone else has other ideas we can incorporate.

For now, though, I’ll stick with the triangle.

The hypotenuse lies flat on the print bed, the adjacent side nestles against the part, and the opposite side has no other function—which defines the sides.

The base has a variable width but should always be a multiple of the nozzle diameter. For a standard 0.4 nozzle, that’s 0.8 mm—which is sufficient for good support.
A surface should be automatically generated around the hypotenuse —elliptical in shape— to secure the triangle’s rather narrow base to the print bed.
A radius as a transition between the triangle and the base plate provides additional stability and just looks good.
Also the Top angle of the Triangle should be rounded and the angle between hypotenuse and the adjacent side.

The distance between the adjacent side and the part should be half the nozzle diameter—in this example, 0.2 mm.

Now we have a nice support structure in the shape of a triangle on one side of our part. However, the triangle isn’t in direct contact with the part, since we want to be able to remove it easily later on. This also means the part can tip over to the other side.
To prevent this, we need small connectors between the triangle and the part.
These connectores are always horizontal and they have to be always exactly at the height of a single print layer.
They are also always exactly one layer high and one nozzle width wide. In this example, that means 0.2 mm high and 0.4 mm wide.

I recommend the following video for more details. from slant3d

Here's a small example contruction
thumbnail.gz
This is the layerling-File, I just could add *.gz-Files. You have to repack ist to zip/lyl
I think, we need the following parameters from the user

1. Nozzle-Diameter
2. Layer height, maybe with a different height of the first layer, to set the small connectors exactly to the right height. Otherwise the slicer might mess with it. We have to give it a try.

If we decide to manually place the support, it should snap to the supported structure, otherwise we need a information where to place it.

That's a lot of text and information, so I'm sure there are some questions.
If I can support you, just ask. You should have my personal E-Mail-Adress, and it's totally ok, if you e-Mail me.

Gruß
Ingo

[uninteressant2026]

I think, it's the slicer respectivly the geometry of the cube. The very first layer of the cube is to fine do get recognized by the slicer.
After putting a small chamfer to the bottom edge it works fine.
Now I'll test the printing.

[henmedia]

@uninteressant2026 good to know, thanks, Ingo. That matches my guess: standing on an edge, the first layer is only a thin strip, and the slicer drops it. A small chamfer on that edge is the right fix, because it gives the part a real foot on the plate. Good luck with the print.

[uninteressant2026]

next little problem. As written I curved the bottem Edge. I can't place a new support to a curved face although it's just a curved edge.
I removed the curved edge, placed the cube on the workplane again, added a support, chamfered the edge again, placed the cube on the workplane again and then i have to move the support to the side because it touched the cube, that's not the best way to do it I think :-)

[henmedia]

@uninteressant2026 thanks, Ingo, that was a real gap, and there was a second one behind it. The rib wanted a flat face that reaches all the way down to the plate, and a rounding ends the face a little above it. Next to a chamfer it was worse: layerling took the flat face for part of a round wall and called it curved. Both are fixed now. Click the flat face above the rounding or chamfer: it may end up to 3 mm above the plate, the rib still goes all the way down, and its bridges start above the rounding.

One more thing: rounding or chamfering the edge a part stands on lifts it a little off the plate. Press D to set it down before you add the rib, otherwise the rib tool asks you to. It's in the next release.

[henmedia]

@uninteressant2026 the support rib on a rounded or chamfered edge is in layerling 1.150.0, out now.