PrusaSlicer bug slicing Pela Block socket panel?  

RE: PrusaSlicer bug slicing Pela Block socket panel?

It points to a bad model... Just because Openscad produces an STL file, that does not guarantee it is printable.

I am sure it is a valid STL file. It is just not printable. A lot of this stuff is iterative. Find what does not work - and fix it.

A good half the STL files I download from Thingiverse have to be fixed, one way or another.

The ukulele I just did had one fret floating in the air above the neck. Looked good - but unprintable.

RE: PrusaSlicer bug slicing Pela Block socket panel?

I looked at Meshmixer and at NetFabb's report.  The mesh appears to be clean by both. But the fact remeshing it allows slicer to work hints at something is different in that cell.  You might want to post it as a defect in Prusa Slicer over on Github; maybe they can chase down what is troubling the slice.  Corner cases are always good for regression testing, too.

RE: PrusaSlicer bug slicing Pela Block socket panel?

3D Builder definitely found a defect in the model - and fixed it.

Unfortunately - I can find no way to force 3D Builder to tell me what it did not like, or how it fixed it.

It is my go-to tool for fixing model issues these days. It is bog simple to use - just have to punch Fix button.

RE: PrusaSlicer bug slicing Pela Block socket panel?

Robert, did 3D builder flag an error, or did you blindly force a repair/remesh?   Even p.slicer doesn't see an error, and since many other tools are not reporting an error, it's hard to blame the model.

RE: PrusaSlicer bug slicing Pela Block socket panel?

3D Builder found the error.

Let me see if this works...

RE: PrusaSlicer bug slicing Pela Block socket panel?

I looked at each of the triangles in question, as far as where the slice goes awry, they all looked okay to my eyeball.  Each one connected.  But we are dealing with E-16 numbers approximating zero, so pretty sure it's an infinitesimal error ten decimal points to the right of E-16.

RE: PrusaSlicer bug slicing Pela Block socket panel?

I had the very same issue with my openscad project and it was all bad counting, there was a nonzero gap between parts of the model. As soon as I fixed my code, problem wanished. As I aim for comparable target, I know that it easy to have code that produces invalid stl only in very rare circumstances while 99% is perfectly sliceable - I had a brick that was half normal and half mass holes and missing mass for example. One layer brick was okay, 3 layer brick was okay, just two layed model was bad... So  ? I guess you found this one in pela bricks.

RE: PrusaSlicer bug slicing Pela Block socket panel?

The problem almost certainly lies in the openSCAD file, and it's not your fault. nor is it an OpenSCAD error..!.

The issue arises when computed differences (one shape cutting bits out of another) come to values that differ in the umpteenth decimal place leaving an impossible to print micro-thin, invisible, remnant. The workaround when this happens is to make the cut-out bigger than it needs to be so that it fully penetrates.

This is a physical manifestation of the difficulty in translating from one number base to another, some decimal values cannot easily be represented in binary.

So increment the cut depth by a tiny amount, render and slice again. If the problem persists come back and let us take a look at the openSCAD script.

RE: PrusaSlicer bug slicing Pela Block socket panel?

I fixed my script and it works fine. My trouble was, that the original work I went from did not use centered objects, so all transformations had to cope with that and translation magic did no always work as it should. Later I found out that I can center everything (I use OpenSCAD for about a month, so learning everything), found out that openscad developers insist on not allowing a way for scripts defaulting to that, so I should end all my lines with, "true" or make wraparounds. This state persist for ten years and while developers tell they are looking for feasible way to implement it, nobody believes that. I decided to make my fork as code where every line ends with the same string is hard to debug.

After centering rewrite, all troubles dissaperared as all transformations are clean and straightforward. I was able to drop around 20% of code...

RE: PrusaSlicer bug slicing Pela Block socket panel?

OpenSCAD is a functional programming language so it's mathematically based and doesn't fit well with the imperative programming habits that many are used to.  You don't have to centre everything but you do have to work within a consistent frame of reference.

It may help to look up 'functional programming' and perhaps follow a course to help grasp the underlying concepts.

Cheerio,

RE: PrusaSlicer bug slicing Pela Block socket panel?

wow, functiuonal programming is here for decades and I worked with some of the environments already - hey, I remember assembler beeing common language to create things, so I have seen very much and see how object programming destroyed everything with whists and wangles of salvation. Everyone can produce any code, much of it ends in working environments and many common tasks take even longer than 20 years ago as result.

No, you have to work in a way, that is consistent with your understanding of reality and perspective, in a way that does not consume time itself without anything created. Some feel it natural that every kind of object starts at random places, are happy with it and feel it much easier than consistent approach. Some dont need it as openscad representation is only result of serious analitical geometry and one transformation more or less with every object make no big difference. Some of us on the other hand need it tidy. I dont even use paper for notes, I do all the transformations in my head and since I went with consistent approach, I rarely have to change the code. So, result is 20% shorter, coded in one third of the time, much less buggy, much easier to debug and much faster to render - do you really think any course could make me produce the longer and less reality based way faster and better? I doubt that.

Yes my code is not heavily functional as that would only slow down the development. I will do it eventualy when I reach the need for libraries, I know it should be coded differently, but the way one passes variables make zero change in the result and I need the result.