faulty threads after slicing
Locked
faulty threads after slicing
 
|
Re: faulty threads after slicing
Administrator
|
Re: faulty threads after slicing
|
|
Re: faulty threads after slicing
|
Unfortunately it looks like threadlib had degenerate faces in the bolt.
and also on the other end. With 2019.05 your model gets a CGAL error. With 2021.01.03 it doesn't, but the export has degenerate faces.
and also on the other side. > -----Original Message----- > From: Discuss [mailto:[hidden email]] On Behalf Of RajHid > Sent: Wed, 6 Jan 2021 09:37 > To: [hidden email] > Subject: Re: [OpenSCAD] faulty threads after slicing > > The Variables anz_x and anz_y determine the amount of threaded holes > one each will result in a Single threaded hole that will do fine on slicing > More will cause problems. > > I also tested to generate two threads by using translate(){} two times whit > same result. > I tested whit ordinary cylinders that will work like a charm. > > I am courious whats wrong. > dosen't make sense. > I think the source is the library. > > greetings. R > > > /* > Generates an Disc shaped insert for a metal box to stor 3D-Printer Nozzles > */ > > // ================================== > // = Used Libraries = > // ================================== > // Used Librareis and usage > // == Threads == > > use <threadlib/threadlib.scad> > // LINK: > //// https://github.com/adrianschlatter/threadlib > // USAGE: > //// bolt("M6", turns=10, higbee_arc=30); > //// tap , bolt, nut > > //m6(); > > module m6(){ > union(){ > difference(){ > cylinder(h=5,d1=25,d2=25, $fn=4); > bolt("M6", turns=6, higbee_arc=30); // tap , bolt, nut > translate([12.5,0,5]){ > //cube([25,25,10],center=true); > } > } > } > } > > > // ================================== > // = Variables = > // ================================== > > // == Disc insert == > > Diameter=51.2; // Disc diameter > Height=10; // Disc Height > > anz_x=2; // count of x holes > anz_y=2; // count of y holes > Stepping=1.5; // amont of distance between holes in hole > diameter > > // === 1 === > > > // === Nozzle === > NozzleThread=6; // Tread Major diameter > NozzlHex_h=3; // Height of hex section > NozzlHex_d=8; // Diameter of hex section > NozzlCone_h=2; // height of the end Cone > NozzlConeBase_d=4.00; // Diameter at base of Cone > NozzlConeTip_d=1.00; // Diameter at tip of Cone > NozzlFilam_d=2; // inner Nozzle diemeter > NozzlFilam_h=10; // lenght of the bore for the filament > NozzlNozzl_d=0.4; // Nozzle diameter for printing > NozzlTreadSection_h=7.50; // Length of the thread section > NozzlTreadSection_minor_d=4.75; // Tread Core Diameter > > NozzlInnerCone_h=2; // > NozzlInnerCone_d=NozzlFilam_d; > > // === 3 === > > // == Positioning Variables == > > // == Dependant Variables == > > > > // === Facettes Numbers === > // - $fn Values for Cylinders and Spheres > > FN_HexNut=6; > FN_Rough=12; > FN_Medium=36; > FN_Fine=72; > > FN_M8=8.00; > > > // == Calculated Variables > > // === 1 === > > // === 2 === > > // === 3 === > > // ================================== > // = Tuning Variables = > // ================================== > // Variables for finetuning (The Slegehammer if something has to be made > fit) > > tune_ScaleThread=0; // widens the M6 Thread of the Nozzle thread > for easy fitting > > // ================================== > // = Test Stage = > // ================================== > > //Holes(); > > // == See me == > // For Testing and Development > > see_me_half(); // for debugging and stuff > module see_me_half(){ > difference(){ > union(){ > Disc_Insert(); > translate([0,0,0]){ > //Nozzle_Template(); > } > } > translate([5,5,0]){ > cube([20,20,50]); // Cutting Cube > } > > } > } > > > // == Cutting == > > // ================================== > // = > // ================================== > > // ================================== > // = Stage = > // ================================== > // Final module for Produktion > > > // > =============================================================================== > // =--------------------------------- Modules > -----------------------------------= > // > =============================================================================== > > module Disc_Insert(){ > difference(){ > cylinder(h=Height,d1=Diameter, d2=Diameter); > translate([ -NozzleThread*Stepping*anz_x/2+NozzleThread*Stepping/2, > -NozzleThread*Stepping*anz_y/2+NozzleThread*Stepping/2, > 0 ]){ > //debug_Holes(); > Holes(); > } > } > } > module Nozzle_Template(){ // a Template of a Nozzle > difference(){ > union(){ > bolt("M6", turns=6, higbee_arc=30); > translate( [0,0,0]){ cylinder( > h=NozzlTreadSection_h, > d1=NozzlTreadSection_minor_d, > d2=NozzlTreadSection_minor_d, > $fn=FN_Fine); > translate( [0,0,NozzlTreadSection_h]){ cylinder( > h=NozzlHex_h, > d1=NozzlHex_d, > d2=NozzlHex_d, > $fn=FN_HexNut); > translate( [0,0,NozzlHex_h]){ cylinder( > h=NozzlCone_h, > d1=NozzlConeBase_d, > d2=NozzlConeTip_d, > $fn=FN_Fine); > } > } > } > } > translate([0,0,NozzlFilam_h+NozzlInnerCone_h]){ > > #cylinder(h=NozzlInnerCone_h,d1=NozzlNozzl_d,d2=NozzlNozzl_d,$fn=FN_Fine); > translate([0,0,0]){ > > //cylinder(h=NozzlFilam_h,d1=NozzlFilam_d,d2=NozzlFilam_d,$fn=FN_Fine); > } > } > translate([0,0,NozzlFilam_h]){ > > #cylinder(h=NozzlInnerCone_h,d1=NozzlFilam_d,d2=NozzlNozzl_d,$fn=FN_Fine); > } > translate([0,0,0]){ > > #cylinder(h=NozzlFilam_h,d1=NozzlFilam_d,d2=NozzlFilam_d,$fn=FN_Fine); > } > translate([0,0,-0.5]){ > cube([10,10,1],center=true); > } > } > } > > // > =============================================================================== > // ---------------------------------- Cutting Modules > ---------------------------- > // > =============================================================================== > > > > //Holes(); > > module debug_Holes(){ // generates two holes whitout using a loop > translate([0,20,0]){ > Thread_Nozzle(); > } > translate([0,-20,0]){ > Thread_Nozzle(); > } > > } > module Holes(){ // generates an array of anz_x times anz_y > objekts > difference(){ > union(){ > for(j=[0:1:anz_y-1]){ > for(i=[0:1:anz_x-1]){ > translate([ NozzleThread*Stepping*i, > 0, > 0 ]){ > //echo(i); > translate([ 0, > NozzleThread*Stepping*j, > 0 ]){ > > //cylinder(h=12,d1=NozzleThread,d2=NozzleThread); > > scale([ 1+tune_ScaleThread, > 1+tune_ScaleThread, > 1 ]){ > translate([0,0,Height]){ > > cylinder(h=Height,d1=NozzleThread*1.2,d2=NozzleThread*1.2,center=true,$fn=FN_Fine); > } > //bolt("M6", turns=15, higbee_arc=30); > Thread_Nozzle(); > } > } > }//NozzleThread*Stepping:Stepping*anz > } > } > } > } > } > > > > module Thread_Nozzle(){ > difference(){ > bolt("M8", turns=15, higbee_arc=30); > cube(10); > } > } > // > =============================================================================== > // ---------------------------------- Linear Extrude Modules > --------------------- > // > =============================================================================== > > > > -- > Sent from: http://forum.openscad.org/ > > _______________________________________________ > OpenSCAD mailing list > http://lists.openscad.org/mailman/listinfo/discuss_lists.openscad.org _______________________________________________ OpenSCAD mailing list [hidden email] http://lists.openscad.org/mailman/listinfo/discuss_lists.openscad.org |
Re: faulty threads after slicing
|
|
Re: faulty threads after slicing
|
Administrator
|
It happens with one thread too
That is Netfabb. A degenerate triangle is tiny, needle like, being so small means that the points can be affected by floating point inaccuracy. They can make a hole in the mesh. The triangle could face the wrong way, ie inside-out. How particular software copes or not is not easily predictable. How you described it: "After sicing the the body and the Threaded holes got invertet so
that there are an array of threaded bolts and not a body of threaded holes." Seems that Prusa followed a hole and ended up treating the inside as
the outside. Netfabb can fix these sometimes. I've never been able to fix them with MeshLab. > -----Original Message----- > From: Discuss [mailto:[hidden email]] On Behalf Of RajHid > Sent: Thu, 7 Jan 2021 21:39 > To: [hidden email] > Subject: Re: [OpenSCAD] faulty threads after slicing > > OK. thanks > > So how do one solve the issue? > And why one Thread will be generated fine and more dosen't > > what kind of Program you were using. > I have MeshLab available and what i have to do? > > > > > -- > Sent from: http://forum.openscad.org/ > > _______________________________________________ > OpenSCAD mailing list > http://lists.openscad.org/mailman/listinfo/discuss_lists.openscad.org _______________________________________________ OpenSCAD mailing list [hidden email] http://lists.openscad.org/mailman/listinfo/discuss_lists.openscad.org
OpenSCAD Admin - email* me if you need anything, or if I've done something stupid...
* on the Forum, click on my MichaelAtOz label, there is a link to email me. Unless specifically shown otherwise above, my contribution is in the Public Domain; to the extent possible under law, I have waived all copyright and related or neighbouring rights to this work. Obviously inclusion of works of previous authors is not included in the above. |
Re: faulty threads after slicing
|
|
I don't have a solution for this particular issue, but I had something similar which I solved by manually using 'dissolve edges' in Blender to create more straightforward (from a human pov - of course this creates ngons) geometry. After doing that, I could use Blender's 'Make manifold' and 'Fill holes' to clean up my model. This is of course not a generalizable solution (not to mention very time consuming), but that process did make me think about whether it would be possible to automate the most tedious parts of it. If I would have had a tool that could replace all groups of triangles in a mesh in which at least two triangles share at least one edge, and of which all the vertices are coplanar within a customizable epsilon, with the concave hull of those triangles, that would have cut down at least 50% of my work. And I think that could solve this particular thread problem too, or at least make it much more obvious where things are going wrong and how to fix it. Similarly, a tool that could 'snap' vertices to say 0.1 model unit increments would have helped a lot (although I imagine that would only make things worse for the thread example). I haven't really thought about whether there are computationally feasible methods of achieving the above, but does anyone know of tools like this? Or maybe there is a way to influence the OpenSCAD STL exporter so that it doesn't triangulate everything in the first place? (I only started using OpenSCAD a few days ago so maybe this is not possible, I can imagine various reasons for why it might not be...) (And of course I understand why ngons for large values of n are frowned upon in mesh models, it's just that I've found that for 3d printing one wants to make different trade offs as for 3d rendering & animation, and that I prefer to slice my ngons manually in some complex cases). cheers Roel On Fri, Jan 8, 2021 at 2:50 AM MichaelAtOz <[hidden email]> wrote:
_______________________________________________ OpenSCAD mailing list [hidden email] http://lists.openscad.org/mailman/listinfo/discuss_lists.openscad.org |
Re: faulty threads after slicing
|
|
| Free forum by Nabble | Edit this page |