Prusaslicer: Can we have a 'never miss a feature' option?

[yoyo42]

Version

2.1.0+win64

Operating system type + version

Windows 10 Pro

3D printer brand / version + firmware version (if known)

Prusa i3 mk3

Behavior

New feature request:

When slicing models with raised text, especially serif fonts, thin parts of letters or thin lines often don't get output by the slicer at all. Often it would be a better end result to print these lines at a single nozzle width EVEN IF that would make the thinnest bits wider than the original model called for, rather than losing them and leaving gaps.

Can we have an optional setting to ensure that even very thin lines are always printed at a single nozzle width instead of being dropped?

[neophyl]

You mean like detect thin walls ?

[yoyo42]

No, that improves nearly-wide-enough cases but doesn't solve it. What I'm asking for really is an override that says 'even if this wall is _much_ thinner than the extrusion width, never lose it completely'.

[supermerill]

In my fork, i added an other setting for thin wall: it's the minimum width. By decreasing it, you can make it create thin walls for features that are thinner than your nozzle, but they will be extruded at nozzle width*1.05.
It's something like that you request?

^{note: You can download a release to test it.}

[Patronics]

I have another situation where this feature would be useful, as I occasionally need to print small parts that were designed optimized for a specific nozzle diameter, but I don't want to swap out the nozzle for the one part (I've had a rather high rate of damaged heat-block cables when swapping out the nozzle in the past), and it's really frustrating that slicer just "optimizes" them away to nothing since they're thinner than the nozzle width.

So far I've been using my local MakerSpace's printers as a workaround whenever I need to make more of them (especially annoying as they only use 3mm Diameter filament), but that's obviously frustrating when I have a printer sitting idle at home. Has anyone else tested @supermerill 's fork? If it doesn't break anything else perhaps you could make a pull request for it?

Edit: I tried to build your fork to test for myself (As you don't seem to provide a prebuilt MacOS binary), but cmake failed with an error like the ones discussed here, and I don't see an obvious resolution to the issue in that thread.

[faboaic]

Maybe you could try horizontal expansion with a positive value as a workaround?

[supermerill]

@Patronics please don't report problem about my fork here, open / reply to an issue in my repository.