This time OpenCSG is right & CGAL is wrong...

classic Classic list List threaded Threaded
11 messages Options
Reply | Threaded
Open this post in threaded view
|

This time OpenCSG is right & CGAL is wrong...

Dan Zuras 3D


        Folks,

        The following snippet of code is intended to make a
        hollow polygonal toroid.  In this instance, I have
        sliced it in half to show the inside.

        This time around it is OpenCSG that gets it right
        & CGAL that fills the void within.

        Alas, as CGAL is involved in making STLs, the actual
        object also gets filled in.

        I cannot make these things this way.  I have 100
        different versions to make & it would take far too
        long to fabricate on my machine.  Not to mention
        far too much plastic.

        Any suggestions?

        Thanks,

                                Dan


module unitCircle() circle(1,$fn=100);
module unitSquare() square([1,1],center=true);

module stick(s,t) union() {
        translate([-s/2,0,0]) scale([t/2,t/2]) unitCircle();
        translate([s/2,0,0]) scale([t/2,t/2]) unitCircle();
        scale([s,t]) unitSquare();
}

module mul1(n,s,t) for (i=[1:n]) rotate([0,0,i*360/n])
        translate([0,-s/(2*tan(180/n)),0]) stick(s,t);

module entry(n,m,s,t) rotate_extrude(convexity=2,$fn=m)
        translate([s+s/(2*sin(180/n)),s/(2*tan(180/n))+t/2,0]) mul1(n,s,t);

intersection() {
        entry(5,5,10,2);
        translate([50,0,0]) cube([100,100,100],center=true);
}

Reply | Threaded
Open this post in threaded view
|

Re: This time OpenCSG is right & CGAL is wrong...

M.Rule
well, if you're using skeinforge to slice and make gcode, you can set the infill to something like 0.000001 and add some extra shells, which would make a solid object print as if hollow. there might also be another way to construct the torus that works around this bug, but I'm not certain.

On Sun, Oct 23, 2011 at 5:01 PM, Dan Zuras 3D <[hidden email]> wrote:


       Folks,

       The following snippet of code is intended to make a
       hollow polygonal toroid.  In this instance, I have
       sliced it in half to show the inside.

       This time around it is OpenCSG that gets it right
       & CGAL that fills the void within.

       Alas, as CGAL is involved in making STLs, the actual
       object also gets filled in.

       I cannot make these things this way.  I have 100
       different versions to make & it would take far too
       long to fabricate on my machine.  Not to mention
       far too much plastic.

       Any suggestions?

       Thanks,

                               Dan


module unitCircle() circle(1,$fn=100);
module unitSquare() square([1,1],center=true);

module stick(s,t) union() {
       translate([-s/2,0,0]) scale([t/2,t/2]) unitCircle();
       translate([s/2,0,0]) scale([t/2,t/2]) unitCircle();
       scale([s,t]) unitSquare();
}

module mul1(n,s,t) for (i=[1:n]) rotate([0,0,i*360/n])
       translate([0,-s/(2*tan(180/n)),0]) stick(s,t);

module entry(n,m,s,t) rotate_extrude(convexity=2,$fn=m)
       translate([s+s/(2*sin(180/n)),s/(2*tan(180/n))+t/2,0]) mul1(n,s,t);

intersection() {
       entry(5,5,10,2);
       translate([50,0,0]) cube([100,100,100],center=true);
}
_______________________________________________
OpenSCAD mailing list
[hidden email]
http://rocklinux.net/mailman/listinfo/openscad

Reply | Threaded
Open this post in threaded view
|

Re: This time OpenCSG is right & CGAL is wrong...

Whosawhatsis
You'll probably have better luck if you try subtracting one solid toroid from another, rather than extruding a shape that already has a hole in it.

On Sunday, October 23, 2011 at 2:06 PM, M.Rule wrote:

well, if you're using skeinforge to slice and make gcode, you can set the infill to something like 0.000001 and add some extra shells, which would make a solid object print as if hollow. there might also be another way to construct the torus that works around this bug, but I'm not certain.

On Sun, Oct 23, 2011 at 5:01 PM, Dan Zuras 3D <[hidden email]> wrote:


       Folks,

       The following snippet of code is intended to make a
       hollow polygonal toroid.  In this instance, I have
       sliced it in half to show the inside.

       This time around it is OpenCSG that gets it right
       & CGAL that fills the void within.

       Alas, as CGAL is involved in making STLs, the actual
       object also gets filled in.

       I cannot make these things this way.  I have 100
       different versions to make & it would take far too
       long to fabricate on my machine.  Not to mention
       far too much plastic.

       Any suggestions?

       Thanks,

                               Dan


module unitCircle() circle(1,$fn=100);
module unitSquare() square([1,1],center=true);

module stick(s,t) union() {
       translate([-s/2,0,0]) scale([t/2,t/2]) unitCircle();
       translate([s/2,0,0]) scale([t/2,t/2]) unitCircle();
       scale([s,t]) unitSquare();
}

module mul1(n,s,t) for (i=[1:n]) rotate([0,0,i*360/n])
       translate([0,-s/(2*tan(180/n)),0]) stick(s,t);

module entry(n,m,s,t) rotate_extrude(convexity=2,$fn=m)
       translate([s+s/(2*sin(180/n)),s/(2*tan(180/n))+t/2,0]) mul1(n,s,t);

intersection() {
       entry(5,5,10,2);
       translate([50,0,0]) cube([100,100,100],center=true);
}
_______________________________________________
OpenSCAD mailing list
[hidden email]
http://rocklinux.net/mailman/listinfo/openscad

_______________________________________________
OpenSCAD mailing list
[hidden email]
http://rocklinux.net/mailman/listinfo/openscad

Reply | Threaded
Open this post in threaded view
|

Re: This time OpenCSG is right & CGAL is wrong...

Dan Zuras 3D
> Date: Sun, 23 Oct 2011 14:27:04 -0700
> From: Whosawhatsis <[hidden email]>
> To: [hidden email]
> Cc: Dan Zuras 3D <[hidden email]>
> Subject: Re: [OpenSCAD] This time OpenCSG is right & CGAL is wrong...
>
>
> You'll probably have better luck if you try subtracting one solid toroid
> from another, rather than extruding a shape that already has a hole in it.

        This is a good suggestion.  Indeed, I used to do that
        when I found that OpenCSG messed up & CGAL did not.
        I tried this new approach in an effort to avoid THAT
        bug.  But given that this one is worse, I may be forced
        to go back to that.  Anyone have any other ideas?
        Or is it possible that the bug is something so simple
        that may be fixed soon?  I have a few weeks but only
        a few.  Thanks.  - Dan

>
> On Sunday, October 23, 2011 at 2:06 PM, M.Rule wrote:
>
> > well, if you're using skeinforge to slice and make gcode, you can set the infill to something like 0.000001 and add some extra shells, which would make a solid object print as if hollow. there might also be another way to construct the torus that works around this bug, but I'm not certain.
> >
> > On Sun, Oct 23, 2011 at 5:01 PM, Dan Zuras 3D <[hidden email] (mailto:[hidden email])> wrote:
> > >
> > >
> > > Folks,
> > >
> > > The following snippet of code is intended to make a
> > > hollow polygonal toroid.  In this instance, I have
> > > sliced it in half to show the inside.
> > >
> > > This time around it is OpenCSG that gets it right
> > > & CGAL that fills the void within.
> > >
> > > Alas, as CGAL is involved in making STLs, the actual
> > > object also gets filled in.
> > >
> > > I cannot make these things this way.  I have 100
> > > different versions to make & it would take far too
> > > long to fabricate on my machine.  Not to mention
> > > far too much plastic.
> > >
> > > Any suggestions?
> > >
> > > Thanks,
> > >
> > >                         Dan
> > >

Reply | Threaded
Open this post in threaded view
|

Re: This time OpenCSG is right & CGAL is wrong...

nophead
OpenCSG doesn't get it right either on my system. It shows daylight through the cut ends instead of the inside of the object.

I find OpenCSG gets a lot of things wrong like that.

On 23 October 2011 22:45, Dan Zuras 3D <[hidden email]> wrote:
> Date: Sun, 23 Oct 2011 14:27:04 -0700
> From: Whosawhatsis <[hidden email]>
> To: [hidden email]
> Cc: Dan Zuras 3D <[hidden email]>
> Subject: Re: [OpenSCAD] This time OpenCSG is right & CGAL is wrong...
>
>
> You'll probably have better luck if you try subtracting one solid toroid
> from another, rather than extruding a shape that already has a hole in it.

       This is a good suggestion.  Indeed, I used to do that
       when I found that OpenCSG messed up & CGAL did not.
       I tried this new approach in an effort to avoid THAT
       bug.  But given that this one is worse, I may be forced
       to go back to that.  Anyone have any other ideas?
       Or is it possible that the bug is something so simple
       that may be fixed soon?  I have a few weeks but only
       a few.  Thanks.  - Dan

>
> On Sunday, October 23, 2011 at 2:06 PM, M.Rule wrote:
>
> > well, if you're using skeinforge to slice and make gcode, you can set the infill to something like 0.000001 and add some extra shells, which would make a solid object print as if hollow. there might also be another way to construct the torus that works around this bug, but I'm not certain.
> >
> > On Sun, Oct 23, 2011 at 5:01 PM, Dan Zuras 3D <[hidden email] (mailto:[hidden email])> wrote:
> > >
> > >
> > > Folks,
> > >
> > > The following snippet of code is intended to make a
> > > hollow polygonal toroid.  In this instance, I have
> > > sliced it in half to show the inside.
> > >
> > > This time around it is OpenCSG that gets it right
> > > & CGAL that fills the void within.
> > >
> > > Alas, as CGAL is involved in making STLs, the actual
> > > object also gets filled in.
> > >
> > > I cannot make these things this way.  I have 100
> > > different versions to make & it would take far too
> > > long to fabricate on my machine.  Not to mention
> > > far too much plastic.
> > >
> > > Any suggestions?
> > >
> > > Thanks,
> > >
> > >                         Dan
> > >
_______________________________________________
OpenSCAD mailing list
[hidden email]
http://rocklinux.net/mailman/listinfo/openscad

Reply | Threaded
Open this post in threaded view
|

Re: This time OpenCSG is right & CGAL is wrong...

William Adams
I would do this:
 
module test_toroid()

{

    difference()
    {
        import_stl("tor1.stl", convexity=5);

        import_stl("tor2.stl", convexity=5);


        translate([-30, 0, -10])

        cube(size=[60, 30, 20]);

    }


    //sor_torus(innerRadius = 15, size=[10,6]); // tor1.stl

    //sor_torus(innerRadius = 15, size=[8,4]); // tor2.stl

}

 

 

In this particular case, I've generated the torus directly, and save it to two .stl files, and then just diff.

 

I so happened to use my own torus generation code: http://www.thingiverse.com/thing:9779

 

But, I don't think it matters.  I think this is the method Dan was using before?  It seems to be the easiest, most reliable.

 

-- William

===============================
- Shaping clay is easier than digging it out of the ground.

http://blog.nanotechstyles.com
http://www.thingiverse.com/WilliamAAdams
https://github.com/Wiladams
 

 


Date: Sun, 23 Oct 2011 22:51:09 +0100
From: [hidden email]
To: [hidden email]; [hidden email]
Subject: Re: [OpenSCAD] This time OpenCSG is right & CGAL is wrong...

OpenCSG doesn't get it right either on my system. It shows daylight through the cut ends instead of the inside of the object.

I find OpenCSG gets a lot of things wrong like that.

On 23 October 2011 22:45, Dan Zuras 3D <[hidden email]> wrote:
> Date: Sun, 23 Oct 2011 14:27:04 -0700
> From: Whosawhatsis <[hidden email]>
> To: [hidden email]
> Cc: Dan Zuras 3D <[hidden email]>
> Subject: Re: [OpenSCAD] This time OpenCSG is right & CGAL is wrong...
>
>
> You'll probably have better luck if you try subtracting one solid toroid
> from another, rather than extruding a shape that already has a hole in it.

       This is a good suggestion.  Indeed, I used to do that
       when I found that OpenCSG messed up & CGAL did not.
       I tried this new approach in an effort to avoid THAT
       bug.  But given that this one is worse, I may be forced
       to go back to that.  Anyone have any other ideas?
       Or is it possible that the bug is something so simple
       that may be fixed soon?  I have a few weeks but only
       a few.  Thanks.  - Dan

>
> On Sunday, October 23, 2011 at 2:06 PM, M.Rule wrote:
>
> > well, if you're using skeinforge to slice and make gcode, you can set the infill to something like 0.000001 and add some extra shells, which would make a solid object print as if hollow. there might also be another way to construct the torus that works around this bug, but I'm not certain.
> >
> > On Sun, Oct 23, 2011 at 5:01 PM, Dan Zuras 3D <[hidden email] (mailto:[hidden email])> wrote:
> > >
> > >
> > > Folks,
> > >
> > > The following snippet of code is intended to make a
> > > hollow polygonal toroid.  In this instance, I have
> > > sliced it in half to show the inside.
> > >
> > > This time around it is OpenCSG that gets it right
> > > & CGAL that fills the void within.
> > >
> > > Alas, as CGAL is involved in making STLs, the actual
> > > object also gets filled in.
> > >
> > > I cannot make these things this way.  I have 100
> > > different versions to make & it would take far too
> > > long to fabricate on my machine.  Not to mention
> > > far too much plastic.
> > >
> > > Any suggestions?
> > >
> > > Thanks,
> > >
> > >                         Dan
> > >
_______________________________________________
OpenSCAD mailing list
[hidden email]
http://rocklinux.net/mailman/listinfo/openscad


_______________________________________________ OpenSCAD mailing list [hidden email] http://rocklinux.net/mailman/listinfo/openscad
Reply | Threaded
Open this post in threaded view
|

Re: This time OpenCSG is right & CGAL is wrong...

Dan Zuras 3D

        Thank you William.

        Alas, the goal of the exercise is not to produce polygonal
        tori but toroidal polygons.

        You see, I was recently asked to help tutor a 4-year old
        in math.  He is not very good with his numbers but he seems
        to have an innate sense of the visual & is very tactile.
        So I thought I'd try to make a 3D multiplication table for
        him.

        The idea is to represent the product MxN by a toroidal
        polygon with M segments on the major axis & N on the
        minor.  Thus looking at either faces or vertices you would
        have MxN in total.

        I want to make 100 of them for the full multiplication
        table with each entry about 2 inches across (so that it
        cannot be swallowed).  There are some special cases for
        1xN, 2xN, Mx1, & Mx2 but I think I have those worked out.

        My experience is that hollow objects of this size &
        complexity come out to 3 or 4 hours each on the machine.
        Were they solid, that 300 or 400 hours would easily
        exceed 1000.  Maybe 2000.  So this is no small project.

        Hell, I have no experience with 4-year olds at all so
        the whole idea may crash & burn.  But I'd like to give
        it a shot.

        So far, short of waiting until OpenSCAD releases a
        version without the bug in it, I think my best bet is
        to subtract an inner toroid from an outer one as
        Whosawhatsis suggested.

        Any other ideas?

        From anyone?

        Thanks all,

                                Dan


> From: William Adams <[hidden email]>
> To: <[hidden email]>, <[hidden email]>
> Subject: RE: [OpenSCAD] This time OpenCSG is right & CGAL is wrong...
> Date: Mon, 24 Oct 2011 02:30:56 +0000
>
>
>
> I would do this: module test_toroid()
> {
>     difference()
>     {
>  import_stl("tor1.stl"=2C convexity=5)=3B        import_stl("tor=
> 2.stl"=2C convexity=5)=3B
>  translate([-30=2C 0=2C -10])        cube(size=[60=2C 30=2C 20])=
> =3B    }
>     //sor_torus(innerRadius = 15=2C size=[10=2C6])=3B // tor1.stl
>     //sor_torus(innerRadius = 15=2C size=[8=2C4])=3B // tor2.stl}  In t=
> his particular case=2C I've generated the torus directly=2C and save it to =
> two .stl files=2C and then just diff. I so happened to use my own torus gen=
> eration code: http://www.thingiverse.com/thing:9779 But=2C I don't think it=
>  matters.  I think this is the method Dan was using before?  It seems to be=
>  the easiest=2C most reliable. -- William
>
> ==========================
> ======
>
> - Shaping clay is easier than digging it out of the ground.
>
>
> http://blog.nanotechstyles.com
> http://www.thingiverse.com/WilliamAAdams
> https://github.com/Wiladams
>

Reply | Threaded
Open this post in threaded view
|

Re: This time OpenCSG is right & CGAL is wrong...

William Adams
Oh, that sounds like a great app.  I was using the simplified form of the torus.  The more
detailed for is the following:
 
 
rotate([0, -90,0])
surface_rotation_torus(offset=[0,15],
size = [10,10],
anglesteps = 5,
sweepsteps = 5);
 
rotate([0, -90,0])
surface_rotation_torus(offset=[0,15],
size = [8,8],
anglesteps = 5,
sweepsteps = 5);

 
The anglesteps, controls how many sides to the small circles.  In this case 5.
The sweepsteps controls how many around the greater circle, in this case, also 5.
 
So, the cross section is a pentagon, and if you look at it from the top, it's also a pentagon.
 
Given your system, the multiple should be 25?
 
And if you wanted 30, you'd just change either of the values to 6.
 
But, that's just this particular method. 
 
The basic idea was to take a shell, subtract a smaller version of it, and you're good to go.
 
-- William
===============================
- Shaping clay is easier than digging it out of the ground.

http://blog.nanotechstyles.com
http://www.thingiverse.com/WilliamAAdams
https://github.com/Wiladams
 

 
> To: [hidden email]

> CC: [hidden email]; [hidden email]
> From: [hidden email]
> Subject: Re: [OpenSCAD] This time OpenCSG is right & CGAL is wrong...
> Date: Sun, 23 Oct 2011 20:16:46 -0700
>
>
> Thank you William.
>
> Alas, the goal of the exercise is not to produce polygonal
> tori but toroidal polygons.
>
> You see, I was recently asked to help tutor a 4-year old
> in math. He is not very good with his numbers but he seems
> to have an innate sense of the visual & is very tactile.
> So I thought I'd try to make a 3D multiplication table for
> him.
>
> The idea is to represent the product MxN by a toroidal
> polygon with M segments on the major axis & N on the
> minor. Thus looking at either faces or vertices you would
> have MxN in total.
>
> I want to make 100 of them for the full multiplication
> table with each entry about 2 inches across (so that it
> cannot be swallowed). There are some special cases for
> 1xN, 2xN, Mx1, & Mx2 but I think I have those worked out.
>
> My experience is that hollow objects of this size &
> complexity come out to 3 or 4 hours each on the machine.
> Were they solid, that 300 or 400 hours would easily
> exceed 1000. Maybe 2000. So this is no small project.
>
> Hell, I have no experience with 4-year olds at all so
> the whole idea may crash & burn. But I'd like to give
> it a shot.
>
> So far, short of waiting until OpenSCAD releases a
> version without the bug in it, I think my best bet is
> to subtract an inner toroid from an outer one as
> Whosawhatsis suggested.
>
> Any other ideas?
>
> >From anyone?
>
> Thanks all,
>
> Dan
>
>
> > From: William Adams <[hidden email]>
> > To: <[hidden email]>, <[hidden email]>
> > Subject: RE: [OpenSCAD] This time OpenCSG is right & CGAL is wrong...
> > Date: Mon, 24 Oct 2011 02:30:56 +0000
> >
> >
> >
> > I would do this: module test_toroid()
> > {
> > difference()
> > {
> > import_stl("tor1.stl"=2C convexity=5)=3B import_stl("tor=
> > 2.stl"=2C convexity=5)=3B
> > translate([-30=2C 0=2C -10]) cube(size=[60=2C 30=2C 20])=
> > =3B }
> > //sor_torus(innerRadius = 15=2C size=[10=2C6])=3B // tor1.stl
> > //sor_torus(innerRadius = 15=2C size=[8=2C4])=3B // tor2.stl} In t=
> > his particular case=2C I've generated the torus directly=2C and save it to =
> > two .stl files=2C and then just diff. I so happened to use my own torus gen=
> > eration code: http://www.thingiverse.com/thing:9779 But=2C I don't think it=
> > matters. I think this is the method Dan was using before? It seems to be=
> > the easiest=2C most reliable. -- William
> >
> > ==========================
> > ======
> >
> > - Shaping clay is easier than digging it out of the ground.
> >
> >
> > http://blog.nanotechstyles.com
> > http://www.thingiverse.com/WilliamAAdams
> > https://github.com/Wiladams
> >
Reply | Threaded
Open this post in threaded view
|

Re: This time OpenCSG is right & CGAL is wrong...

Dan Zuras 3D

        Cool.  OK, I will try something
        along these lines.  Thanks.  - Dan

> From: William Adams <[hidden email]>
> To: <[hidden email]>
> CC: <[hidden email]>
> Subject: RE: [OpenSCAD] This time OpenCSG is right & CGAL is wrong...
> Date: Mon, 24 Oct 2011 04:04:08 +0000
>
>
> Oh=2C that sounds like a great app.  I was using the simplified form of the=
>  torus.  The more detailed for is the following:  rotate([0=2C -90=2C0])
> surface_rotation_torus(offset=[0=2C15]=2C
> size = [10=2C10]=2C
> anglesteps = 5=2C
> sweepsteps = 5)=3B rotate([0=2C -90=2C0])
> surface_rotation_torus(offset=[0=2C15]=2C
> size = [8=2C8]=2C
> anglesteps = 5=2C
> sweepsteps = 5)=3B
>  The anglesteps=2C controls how many sides to the small circles.  In this c=
> ase 5.The sweepsteps controls how many around the greater circle=2C in this=
>  case=2C also 5. So=2C the cross section is a pentagon=2C and if you look a=
> t it from the top=2C it's also a pentagon. Given your system=2C the multipl=
> e should be 25? And if you wanted 30=2C you'd just change either of the val=
> ues to 6. But=2C that's just this particular method.   The basic idea was t=
> o take a shell=2C subtract a smaller version of it=2C and you're good to go=
> . -- William

Reply | Threaded
Open this post in threaded view
|

Re: This time OpenCSG is right & CGAL is wrong...

Whosawhatsis
If you're just making it hollow to decrease printing time, I will second the suggestion to just use skeinforge to make it hollow. You can set the infill to 0 and give it a few extra shells to make the walls thick enough for strength. The object will print faster that way, and this also has the advantage of allowing skeinforge to fill some of that space if it needs to because the internal overhangs are too steep, and the object would otherwise cave and cause a failed print.

On Monday, October 24, 2011 at 1:16 AM, Dan Zuras 3D wrote:


Cool. OK, I will try something
along these lines. Thanks. - Dan

From: William Adams <[hidden email]>
To: <[hidden email]>
CC: <[hidden email]>
Subject: RE: [OpenSCAD] This time OpenCSG is right & CGAL is wrong...
Date: Mon, 24 Oct 2011 04:04:08 +0000


Oh=2C that sounds like a great app. I was using the simplified form of the=
torus. The more detailed for is the following: rotate([0=2C -90=2C0])
surface_rotation_torus(offset=[0=2C15]=2C
size = [10=2C10]=2C
anglesteps = 5=2C
sweepsteps = 5)=3B rotate([0=2C -90=2C0])
surface_rotation_torus(offset=[0=2C15]=2C
size = [8=2C8]=2C
anglesteps = 5=2C
sweepsteps = 5)=3B
The anglesteps=2C controls how many sides to the small circles. In this c=
ase 5.The sweepsteps controls how many around the greater circle=2C in this=
case=2C also 5. So=2C the cross section is a pentagon=2C and if you look a=
t it from the top=2C it's also a pentagon. Given your system=2C the multipl=
e should be 25? And if you wanted 30=2C you'd just change either of the val=
ues to 6. But=2C that's just this particular method. The basic idea was t=
o take a shell=2C subtract a smaller version of it=2C and you're good to go=
. -- William
_______________________________________________
OpenSCAD mailing list
[hidden email]
http://rocklinux.net/mailman/listinfo/openscad

Reply | Threaded
Open this post in threaded view
|

Re: This time OpenCSG is right & CGAL is wrong...

Dan Zuras 3D
> Date: Mon, 24 Oct 2011 04:50:27 -0700
> From: Whosawhatsis <[hidden email]>
> To: Dan Zuras 3D <[hidden email]>, [hidden email]
> Cc: William Adams <[hidden email]>
> Subject: Re: [OpenSCAD] This time OpenCSG is right & CGAL is wrong...
>
>
> If you're just making it hollow to decrease printing time, I will
> second the suggestion to just use skeinforge to make it hollow.
> You can set the infill to 0 and give it a few extra shells to make
> the walls thick enough for strength. The object will print faster
> that way, and this also has the advantage of allowing skeinforge
> to fill some of that space if it needs to because the internal
> overhangs are too steep, and the object would otherwise cave and
> cause a failed print.
>

        OK, I have created a skeinforge model that deactivates
        fill.  It has has 4 extra shells on the Alternating
        Solid Layer & 3 each on the Base & Sparse Layers.
        Given that the even entries WILL have flat overhangs,
        are these the correct settings WRT overhang?  - Dan

        (Please excuse the slight basenote drift all.)