constructive-lib funnel?
Not really because someone could put in a library and break client code.
On Fri, 22 Sept 2023, 19:55 pproj@posteo.de, pproj@posteo.de wrote:
OK, i see.
Thanks nophead.
i still think that the "Option Explicit" like mechanism should resolve
the Problem
On 22.09.23 20:46, nop head wrote:
The reason is because the original implementation of openscad had a
limitation that only allowed module to return a single child. That was
found to be a limitation when using children() as the only way to pass
multiple children was by enclosing them in braces. You couldn't use a for
loop or an object made by a user module. Also a big performance hit
doing a
top level implicit union for output formats that support multiple
objects.
The only way to fix it is not backwards compatible.
On Fri, 22 Sept 2023, 18:49 pproj@posteo.de, pproj@posteo.de wrote:
Thanks Nophead,
i meant why, in the sense of what are the reasons for it.
perhaps we could have something like "option explicit" for javascript,
so if one includes it in the code, the code will not compile until you
turn the Lazy unions off.
Resulting in an Error Message, telling what to do.
so everybody could ensure compatibility, but experimental features can
still be enabled
On 22.09.23 18:53, nop head wrote:
Because without lazy unions most operations do an implicit union and
return
a single child. Lazy unions stop that so in some places you would need
to
add an explicit union to preserve the semantics. That is why it is
disabled
by default. I personally never enable it.
On Fri, 22 Sept 2023, 17:47 pproj@posteo.de, pproj@posteo.de wrote:
ehmm... it is indeed a strong semantic change then. do you know why?
On 22.09.23 18:34, nop head wrote:
Because without lazy union a for does an implicit union, so you get
an
intersection with one child that is a NOP. When lazy union is on each
iteration of the loop will be intersected with all the others. There
is
intersection_for to do the same thing without lazy unions.
On Fri, 22 Sept 2023 at 17:16, pproj@posteo.de pproj@posteo.de
wrote:
So Ken has figured out the cause for extrusion problems was "lazy
unions",
i was able to fix it by removing a single child intersection() which
was
not needed any more and doing nothing anyway, from the code below,
Anybody any idea, why a single child intersection() causes problems
with
lazy unions?
[...]
doChamferBox(lx=lx,ly=ly,lz=lz)
cube([lx,ly,lz],center=true);
[.....]
module doChamferBox(lx,ly,lz,chamferInfo=chamferInfo(),childFn=$fn)
{
[.....]
for( i = [chamferInfo[0] , chamferInfo[1]] )
{
r=i[1];
rSide=i[2];
fnCorner=i[3];
up( (lz==undef )?0:((lz/2-abs(r))i[0]))
mirror([0,0,i[0]>0?0:1])
linear_extrude(height=abs(r)
,slices=1
,scale=[(lx+2r)/lx,(ly+2*r)/ly])
resize([lx,ly])
offset(r = abs( rSide ), $fn = fnCorner)
square([lx,ly],center=true,$fn=childFn);
}
//this inteesection() seems is a non functional leftover and was
causing
the problem with lazy unions
intersection()
{
for(i=[chamferInfo[0],chamferInfo[1]])
{
r=i[1];
rSide=i[2];
fnCorner=i[3];
mirror([0,0,i[0]>0?0:1])
down(chamferInfo[1]==undef?lz/2:0.02)
linear_extrude(
height=(chamferInfo[1]==undef?lz:lz/2)
-abs(r)+.05
,slices=1)
resize([lx,ly])
offset(r=abs(rSide),$fn=fnCorner)
square([lx,ly],center=true,$fn=childFn);
}
}
else children();
}
On 22.09.23 00:52, Ken wrote:
include<constructive-compiled.scad>
$fa = 2;
$fs = 0.5;
chamfer(-4,-2) tube(d=20, h=12, wall=3);
it works as expected- I get a nicely chamfered tube. If I
change
it
to
chamfer(-2,-2,-2) g(turnXY(45), X(60)) box(20);
or
chamfer(-1,-2,-3) X(-40)box(10,x=35,h=15);
On 22.09.23 17:07, pproj@posteo.de wrote:
HEllo Ken, thanks for figuring out cause,
there are various ways:
include <../../devlibs/constructive/constructive-all.scad>
$fn=60;
assemble()
{
add()
tube(d=20,d2=10,h=10,wall=3);
//chamfering has a side effect when using d2, so it will look
like
this
add(X(25),chamfer(-2,-2))
tube(d=20,d2=10,h=10,wall=3);
//i would use the bentStrip()with a 2d primitive Bodylike circle,
instead
add(X(50),reflectZ())
rotate_extrude()
X(6)bentStrip([Y(10),turnXY(-45),Y(10)])
circle(d=2);
}
i would need to add this example to the tutorial or the
Constructive's
Gallery
beside this you could always use regular cylinder([]) instead of
tube,
it just will not react on alignments, but it will on translations
like
X(10)
On 22.09.23 01:28, Ken wrote:
Is there a way of constructing a funnel using the
constructive-compiled library? I searched the tutorials but didn't
find anything, and have experimented with the syntax of a
chamfered
tube with no success. Normally I would just difference two cones,
but
it would be interesting to do it using theconstructive-compiled
library..
OpenSCAD mailing list
To unsubscribe send an email to discuss-leave@lists.openscad.org
OpenSCAD mailing list
To unsubscribe send an email to discuss-leave@lists.openscad.org
OpenSCAD mailing list
To unsubscribe send an email to discuss-leave@lists.openscad.org
OpenSCAD mailing list
To unsubscribe send an email to discuss-leave@lists.openscad.org
OpenSCAD mailing list
To unsubscribe send an email to discuss-leave@lists.openscad.org
OpenSCAD mailing list
To unsubscribe send an email to discuss-leave@lists.openscad.org
OpenSCAD mailing list
To unsubscribe send an email to discuss-leave@lists.openscad.org
OpenSCAD mailing list
To unsubscribe send an email to discuss-leave@lists.openscad.org
OpenSCAD mailing list
To unsubscribe send an email to discuss-leave@lists.openscad.org
Or a keyword prefixing the "Module" like "async" in other languages
specifying that this module it is not unioning,
but i believe this is much more work to implement, than one a simple
assert with "Option Explicit" /lazyOff
On 22.09.23 20:55, pproj@posteo.de wrote:
OK, i see.
Thanks nophead.
i still think that the "Option Explicit" like mechanism should resolve
the Problem
On 22.09.23 20:46, nop head wrote:
The reason is because the original implementation of openscad had a
limitation that only allowed module to return a single child. That was
found to be a limitation when using children() as the only way to pass
multiple children was by enclosing them in braces. You couldn't use a
for
loop or an object made by a user module. Also a big performance hit
doing a
top level implicit union for output formats that support multiple
objects.
The only way to fix it is not backwards compatible.
On Fri, 22 Sept 2023, 18:49 pproj@posteo.de, pproj@posteo.de wrote:
Thanks Nophead,
i meant why, in the sense of what are the reasons for it.
perhaps we could have something like "option explicit" for javascript,
so if one includes it in the code, the code will not compile until you
turn the Lazy unions off.
Resulting in an Error Message, telling what to do.
so everybody could ensure compatibility, but experimental features can
still be enabled
On 22.09.23 18:53, nop head wrote:
Because without lazy unions most operations do an implicit union and
return
a single child. Lazy unions stop that so in some places you would
need to
add an explicit union to preserve the semantics. That is why it is
disabled
by default. I personally never enable it.
On Fri, 22 Sept 2023, 17:47 pproj@posteo.de, pproj@posteo.de wrote:
ehmm... it is indeed a strong semantic change then. do you know why?
On 22.09.23 18:34, nop head wrote:
Because without lazy union a for does an implicit union, so you
get an
intersection with one child that is a NOP. When lazy union is on
each
iteration of the loop will be intersected with all the others.
There is
intersection_for to do the same thing without lazy unions.
On Fri, 22 Sept 2023 at 17:16, pproj@posteo.de pproj@posteo.de
wrote:
So Ken has figured out the cause for extrusion problems was "lazy
unions",
i was able to fix it by removing a single child intersection()
which
was
not needed any more and doing nothing anyway, from the code below,
Anybody any idea, why a single child intersection() causes problems
with
lazy unions?
[...]
doChamferBox(lx=lx,ly=ly,lz=lz)
cube([lx,ly,lz],center=true);
[.....]
module doChamferBox(lx,ly,lz,chamferInfo=chamferInfo(),childFn=$fn)
{
[.....]
for( i = [chamferInfo[0] , chamferInfo[1]] )
{
r=i[1];
rSide=i[2];
fnCorner=i[3];
up( (lz==undef )?0:((lz/2-abs(r))i[0]))
mirror([0,0,i[0]>0?0:1])
linear_extrude(height=abs(r)
,slices=1
,scale=[(lx+2r)/lx,(ly+2*r)/ly])
resize([lx,ly])
offset(r = abs( rSide ), $fn = fnCorner)
square([lx,ly],center=true,$fn=childFn);
}
//this inteesection() seems is a non functional leftover and was
causing
the problem with lazy unions
intersection()
{
for(i=[chamferInfo[0],chamferInfo[1]])
{
r=i[1];
rSide=i[2];
fnCorner=i[3];
mirror([0,0,i[0]>0?0:1])
down(chamferInfo[1]==undef?lz/2:0.02)
linear_extrude(
height=(chamferInfo[1]==undef?lz:lz/2)
-abs(r)+.05
,slices=1)
resize([lx,ly])
offset(r=abs(rSide),$fn=fnCorner)
square([lx,ly],center=true,$fn=childFn);
}
}
else children();
}
On 22.09.23 00:52, Ken wrote:
>
> include<constructive-compiled.scad>
> $fa = 2;
> $fs = 0.5;
>
> chamfer(-4,-2) tube(d=20, h=12, wall=3);
>
> it works as expected- I get a nicely chamfered tube. If I
change
it
to
>
> chamfer(-2,-2,-2) g(turnXY(45), X(60)) box(20);
>
> or
>
> chamfer(-1,-2,-3) X(-40)box(10,x=35,h=15);
On 22.09.23 17:07, pproj@posteo.de wrote:
HEllo Ken, thanks for figuring out cause,
there are various ways:
include <../../devlibs/constructive/constructive-all.scad>
$fn=60;
assemble()
{
add()
tube(d=20,d2=10,h=10,wall=3);
//chamfering has a side effect when using d2, so it will
look like
this
add(X(25),chamfer(-2,-2))
tube(d=20,d2=10,h=10,wall=3);
//i would use the bentStrip()with a 2d primitive Bodylike circle,
instead
add(X(50),reflectZ())
rotate_extrude()
X(6)bentStrip([Y(10),turnXY(-45),Y(10)])
circle(d=2);
}
i would need to add this example to the tutorial or the
Constructive's
Gallery
beside this you could always use regular cylinder([]) instead of
tube,
it just will not react on alignments, but it will on translations
like
X(10)
On 22.09.23 01:28, Ken wrote:
Is there a way of constructing a funnel using the
constructive-compiled library? I searched the tutorials but
didn't
find anything, and have experimented with the syntax of a
chamfered
tube with no success. Normally I would just difference two cones,
but
it would be interesting to do it using theconstructive-compiled
library..
OpenSCAD mailing list
To unsubscribe send an email to discuss-leave@lists.openscad.org
OpenSCAD mailing list
To unsubscribe send an email to discuss-leave@lists.openscad.org
OpenSCAD mailing list
To unsubscribe send an email to discuss-leave@lists.openscad.org
OpenSCAD mailing list
To unsubscribe send an email to discuss-leave@lists.openscad.org
OpenSCAD mailing list
To unsubscribe send an email to discuss-leave@lists.openscad.org
OpenSCAD mailing list
To unsubscribe send an email to discuss-leave@lists.openscad.org
OpenSCAD mailing list
To unsubscribe send an email to discuss-leave@lists.openscad.org
OpenSCAD mailing list
To unsubscribe send an email to discuss-leave@lists.openscad.org
This the same issue I wrote to this discussion list Sept 19, under the
title:
Dev build has different semantics than the 2021.1 release
Specifically, I was using some else’s library and the change broke my code.
On Fri, Sep 22, 2023 at 12:06 PM nop head nop.head@gmail.com wrote:
Not really because someone could put in a library and break client code.
On Fri, 22 Sept 2023, 19:55 pproj@posteo.de, pproj@posteo.de wrote:
OK, i see.
Thanks nophead.
i still think that the "Option Explicit" like mechanism should resolve
the Problem
On 22.09.23 20:46, nop head wrote:
The reason is because the original implementation of openscad had a
limitation that only allowed module to return a single child. That was
found to be a limitation when using children() as the only way to pass
multiple children was by enclosing them in braces. You couldn't use a
for
loop or an object made by a user module. Also a big performance hit
doing a
top level implicit union for output formats that support multiple
objects.
The only way to fix it is not backwards compatible.
On Fri, 22 Sept 2023, 18:49 pproj@posteo.de, pproj@posteo.de wrote:
Thanks Nophead,
i meant why, in the sense of what are the reasons for it.
perhaps we could have something like "option explicit" for javascript,
so if one includes it in the code, the code will not compile until you
turn the Lazy unions off.
Resulting in an Error Message, telling what to do.
so everybody could ensure compatibility, but experimental features can
still be enabled
On 22.09.23 18:53, nop head wrote:
Because without lazy unions most operations do an implicit union and
return
a single child. Lazy unions stop that so in some places you would
need to
add an explicit union to preserve the semantics. That is why it is
disabled
by default. I personally never enable it.
On Fri, 22 Sept 2023, 17:47 pproj@posteo.de, pproj@posteo.de wrote:
ehmm... it is indeed a strong semantic change then. do you know why?
On 22.09.23 18:34, nop head wrote:
Because without lazy union a for does an implicit union, so you get
an
intersection with one child that is a NOP. When lazy union is on
each
iteration of the loop will be intersected with all the others.
There is
intersection_for to do the same thing without lazy unions.
On Fri, 22 Sept 2023 at 17:16, pproj@posteo.de pproj@posteo.de
wrote:
So Ken has figured out the cause for extrusion problems was "lazy
unions",
i was able to fix it by removing a single child intersection()
which
was
not needed any more and doing nothing anyway, from the code below,
Anybody any idea, why a single child intersection() causes problems
with
lazy unions?
[...]
doChamferBox(lx=lx,ly=ly,lz=lz)
cube([lx,ly,lz],center=true);
[.....]
module doChamferBox(lx,ly,lz,chamferInfo=chamferInfo(),childFn=$fn)
{
[.....]
for( i = [chamferInfo[0] , chamferInfo[1]] )
{
r=i[1];
rSide=i[2];
fnCorner=i[3];
up( (lz==undef )?0:((lz/2-abs(r))i[0]))
mirror([0,0,i[0]>0?0:1])
linear_extrude(height=abs(r)
,slices=1
,scale=[(lx+2r)/lx,(ly+2*r)/ly])
resize([lx,ly])
offset(r = abs( rSide ), $fn = fnCorner)
square([lx,ly],center=true,$fn=childFn);
}
//this inteesection() seems is a non functional leftover and was
causing
the problem with lazy unions
intersection()
{
for(i=[chamferInfo[0],chamferInfo[1]])
{
r=i[1];
rSide=i[2];
fnCorner=i[3];
mirror([0,0,i[0]>0?0:1])
down(chamferInfo[1]==undef?lz/2:0.02)
linear_extrude(
height=(chamferInfo[1]==undef?lz:lz/2)
-abs(r)+.05
,slices=1)
resize([lx,ly])
offset(r=abs(rSide),$fn=fnCorner)
square([lx,ly],center=true,$fn=childFn);
}
}
else children();
}
On 22.09.23 00:52, Ken wrote:
include<constructive-compiled.scad>
$fa = 2;
$fs = 0.5;
chamfer(-4,-2) tube(d=20, h=12, wall=3);
it works as expected- I get a nicely chamfered tube. If I
change
it
to
chamfer(-2,-2,-2) g(turnXY(45), X(60)) box(20);
or
chamfer(-1,-2,-3) X(-40)box(10,x=35,h=15);
On 22.09.23 17:07, pproj@posteo.de wrote:
HEllo Ken, thanks for figuring out cause,
there are various ways:
include <../../devlibs/constructive/constructive-all.scad>
$fn=60;
assemble()
{
add()
tube(d=20,d2=10,h=10,wall=3);
//chamfering has a side effect when using d2, so it will look
like
this
add(X(25),chamfer(-2,-2))
tube(d=20,d2=10,h=10,wall=3);
//i would use the bentStrip()with a 2d primitive Bodylike circle,
instead
add(X(50),reflectZ())
rotate_extrude()
X(6)bentStrip([Y(10),turnXY(-45),Y(10)])
circle(d=2);
}
i would need to add this example to the tutorial or the
Constructive's
Gallery
beside this you could always use regular cylinder([]) instead of
tube,
it just will not react on alignments, but it will on translations
like
X(10)
On 22.09.23 01:28, Ken wrote:
Is there a way of constructing a funnel using the
constructive-compiled library? I searched the tutorials but
didn't
find anything, and have experimented with the syntax of a
chamfered
tube with no success. Normally I would just difference two cones,
but
it would be interesting to do it using theconstructive-compiled
library..
OpenSCAD mailing list
To unsubscribe send an email to discuss-leave@lists.openscad.org
OpenSCAD mailing list
To unsubscribe send an email to discuss-leave@lists.openscad.org
OpenSCAD mailing list
To unsubscribe send an email to discuss-leave@lists.openscad.org
OpenSCAD mailing list
To unsubscribe send an email to discuss-leave@lists.openscad.org
OpenSCAD mailing list
To unsubscribe send an email to discuss-leave@lists.openscad.org
OpenSCAD mailing list
To unsubscribe send an email to discuss-leave@lists.openscad.org
OpenSCAD mailing list
To unsubscribe send an email to discuss-leave@lists.openscad.org
OpenSCAD mailing list
To unsubscribe send an email to discuss-leave@lists.openscad.org
OpenSCAD mailing list
To unsubscribe send an email to discuss-leave@lists.openscad.org
OpenSCAD mailing list
To unsubscribe send an email to discuss-leave@lists.openscad.org
i do not think so, at all.
just to make it clear, what i mean with "option explicit":
"use strict" in .js code turns the .js interpreter into Strict mode,
it will complain i.e. if variables are used without declarations.
"use strict" in the .scad library's code would throw an error message if
the Lazy Unions Option or other compatibility breaking options are on.
it tells the user exactly what to do, and if your code is based on lazy
unions, you will see it is broken after turning lazy unions off.
so there is no problem. it gives clear guidance, what to do to make it work.
On 22.09.23 20:58, nop head wrote:
Not really because someone could put in a library and break client code.
On Fri, 22 Sept 2023, 19:55 pproj@posteo.de, pproj@posteo.de wrote:
OK, i see.
Thanks nophead.
i still think that the "Option Explicit" like mechanism should resolve
the Problem
On 22.09.23 20:46, nop head wrote:
The reason is because the original implementation of openscad had a
limitation that only allowed module to return a single child. That was
found to be a limitation when using children() as the only way to pass
multiple children was by enclosing them in braces. You couldn't use a for
loop or an object made by a user module. Also a big performance hit
doing a
top level implicit union for output formats that support multiple
objects.
The only way to fix it is not backwards compatible.
On Fri, 22 Sept 2023, 18:49 pproj@posteo.de, pproj@posteo.de wrote:
Thanks Nophead,
i meant why, in the sense of what are the reasons for it.
perhaps we could have something like "option explicit" for javascript,
so if one includes it in the code, the code will not compile until you
turn the Lazy unions off.
Resulting in an Error Message, telling what to do.
so everybody could ensure compatibility, but experimental features can
still be enabled
On 22.09.23 18:53, nop head wrote:
Because without lazy unions most operations do an implicit union and
return
a single child. Lazy unions stop that so in some places you would need
to
add an explicit union to preserve the semantics. That is why it is
disabled
by default. I personally never enable it.
On Fri, 22 Sept 2023, 17:47 pproj@posteo.de, pproj@posteo.de wrote:
ehmm... it is indeed a strong semantic change then. do you know why?
On 22.09.23 18:34, nop head wrote:
Because without lazy union a for does an implicit union, so you get
an
intersection with one child that is a NOP. When lazy union is on each
iteration of the loop will be intersected with all the others. There
is
intersection_for to do the same thing without lazy unions.
On Fri, 22 Sept 2023 at 17:16, pproj@posteo.de pproj@posteo.de
wrote:
So Ken has figured out the cause for extrusion problems was "lazy
unions",
i was able to fix it by removing a single child intersection() which
was
not needed any more and doing nothing anyway, from the code below,
Anybody any idea, why a single child intersection() causes problems
with
lazy unions?
[...]
doChamferBox(lx=lx,ly=ly,lz=lz)
cube([lx,ly,lz],center=true);
[.....]
module doChamferBox(lx,ly,lz,chamferInfo=chamferInfo(),childFn=$fn)
{
[.....]
for( i = [chamferInfo[0] , chamferInfo[1]] )
{
r=i[1];
rSide=i[2];
fnCorner=i[3];
up( (lz==undef )?0:((lz/2-abs(r))i[0]))
mirror([0,0,i[0]>0?0:1])
linear_extrude(height=abs(r)
,slices=1
,scale=[(lx+2r)/lx,(ly+2*r)/ly])
resize([lx,ly])
offset(r = abs( rSide ), $fn = fnCorner)
square([lx,ly],center=true,$fn=childFn);
}
//this inteesection() seems is a non functional leftover and was
causing
the problem with lazy unions
intersection()
{
for(i=[chamferInfo[0],chamferInfo[1]])
{
r=i[1];
rSide=i[2];
fnCorner=i[3];
mirror([0,0,i[0]>0?0:1])
down(chamferInfo[1]==undef?lz/2:0.02)
linear_extrude(
height=(chamferInfo[1]==undef?lz:lz/2)
-abs(r)+.05
,slices=1)
resize([lx,ly])
offset(r=abs(rSide),$fn=fnCorner)
square([lx,ly],center=true,$fn=childFn);
}
}
else children();
}
On 22.09.23 00:52, Ken wrote:
include<constructive-compiled.scad>
$fa = 2;
$fs = 0.5;
chamfer(-4,-2) tube(d=20, h=12, wall=3);
it works as expected- I get a nicely chamfered tube. If I
change
it
to
chamfer(-2,-2,-2) g(turnXY(45), X(60)) box(20);
or
chamfer(-1,-2,-3) X(-40)box(10,x=35,h=15);
On 22.09.23 17:07, pproj@posteo.de wrote:
HEllo Ken, thanks for figuring out cause,
there are various ways:
include <../../devlibs/constructive/constructive-all.scad>
$fn=60;
assemble()
{
add()
tube(d=20,d2=10,h=10,wall=3);
//chamfering has a side effect when using d2, so it will look
like
this
add(X(25),chamfer(-2,-2))
tube(d=20,d2=10,h=10,wall=3);
//i would use the bentStrip()with a 2d primitive Bodylike circle,
instead
add(X(50),reflectZ())
rotate_extrude()
X(6)bentStrip([Y(10),turnXY(-45),Y(10)])
circle(d=2);
}
i would need to add this example to the tutorial or the
Constructive's
Gallery
beside this you could always use regular cylinder([]) instead of
tube,
it just will not react on alignments, but it will on translations
like
X(10)
On 22.09.23 01:28, Ken wrote:
Is there a way of constructing a funnel using the
constructive-compiled library? I searched the tutorials but didn't
find anything, and have experimented with the syntax of a
chamfered
tube with no success. Normally I would just difference two cones,
but
it would be interesting to do it using theconstructive-compiled
library..
OpenSCAD mailing list
To unsubscribe send an email to discuss-leave@lists.openscad.org
OpenSCAD mailing list
To unsubscribe send an email to discuss-leave@lists.openscad.org
OpenSCAD mailing list
To unsubscribe send an email to discuss-leave@lists.openscad.org
OpenSCAD mailing list
To unsubscribe send an email to discuss-leave@lists.openscad.org
OpenSCAD mailing list
To unsubscribe send an email to discuss-leave@lists.openscad.org
OpenSCAD mailing list
To unsubscribe send an email to discuss-leave@lists.openscad.org
OpenSCAD mailing list
To unsubscribe send an email to discuss-leave@lists.openscad.org
OpenSCAD mailing list
To unsubscribe send an email to discuss-leave@lists.openscad.org
OpenSCAD mailing list
To unsubscribe send an email to discuss-leave@lists.openscad.org
OpenSCAD mailing list
To unsubscribe send an email to discuss-leave@lists.openscad.org
Thanks Peter- looks like I was close, but I was trying d1 & d2 instead
of d &d2. Close, but no cigar!
It simplified down to just these few lines to do the job-
include<constructive-compiled.scad>
$fa = 2;
$fs = 0.5;
//Flange
Z(-1.5)tube(d=85, h=3, wall=10.5);
//Funnel
Z(15)tube(d=70,d2=16,h=30,wall=3);
//Tube
Z(95)tube(d=16, h=130, wall=1.5);
Thanks for all your work- I'm going to enjoy using this library.
On 2023-09-23 01:07, pproj@posteo.de wrote:
HEllo Ken, thanks for figuring out cause,
there are various ways:
include <../../devlibs/constructive/constructive-all.scad>
$fn=60;
assemble()
{
add()
tube(d=20,d2=10,h=10,wall=3);
//chamfering has a side effect when using d2, so it will look like this
add(X(25),chamfer(-2,-2))
tube(d=20,d2=10,h=10,wall=3);
//i would use the bentStrip()with a 2d primitive Bodylike circle, instead
add(X(50),reflectZ())
rotate_extrude()
X(6)bentStrip([Y(10),turnXY(-45),Y(10)])
circle(d=2);
}
i would need to add this example to the tutorial or the Constructive's
Gallery
beside this you could always use regular cylinder([]) instead of tube,
it just will not react on alignments, but it will on translations like
X(10)
On 22.09.23 01:28, Ken wrote:
Is there a way of constructing a funnel using the
constructive-compiled library? I searched the tutorials but didn't
find anything, and have experimented with the syntax of a chamfered
tube with no success. Normally I would just difference two cones, but
it would be interesting to do it using theconstructive-compiled
library..
OpenSCAD mailing list
To unsubscribe send an email to discuss-leave@lists.openscad.org
OpenSCAD mailing list
To unsubscribe send an email todiscuss-leave@lists.openscad.org
--
Cheers, Ken
bats059@gmail.com
https://vk7krj.com
https://vk7krj.com/running.html
A baby can be defined as an ego with a noise at one end and a smell at the other.
Your job as parents is to teach them to control all three.
My job as a grandad is to tell you how you are doing it all wrong!
Hello Ken,
great that we have resolved the issue with Lazy U.,
you might also consider stack() for this. it will calculate the Z()'s
for you.
just be careful not put in any intermediate semicolons, while you are
stacking objects. (it has compilation speed penaltly compared to your
code. but is better reflecting the design intent in the code i think)
include<constructive-compiled.scad>
$fa = 2;
$fs = 0.5;
TOUP()
stack()
tube(d=85, h=3, wall=10.5) //Flange
tube(d=70,d2=16,h=30,wall=3) //Funnel
tube(d=16, h=130, wall=1.5); //Tube
//that is it
//alternatively: the beauty of the bentstrip extrusion, is that it can
Provide a wide variety of forms:
X(70)
rotate_extrude()
X(9)bentStrip([X(-3),Y(20),
turnXY(-45),Y(10),
turnXY(-30),Y(15),
turnXY(60),Y(15),
turnXY(-30),Y(20),X(8),Y(10)])
circle(d=2);
one more side note on using bentStrip(),
to avoid openscad compiler getting very very slow, we need to avoid the
steps in which the circle()used to cleate a strip is only rotated, but
not moved (i mean translated by that).
to do so we need to combine every turn step with the following movement
step using asterisk, so this combined operations will be done in a
single step.
here is what i mean:
include<constructive-compiled.scad>
rotate_extrude()
X(9)bentStrip([X(-3),Y(20),
turnXY(-45),Y(10),
turnXY(-30),Y(15),
turnXY(60),Y(15),
turnXY(-30),Y(20),X(8),Y(10)])
circle(d=2);
// needs to become: (it does just the same, but compiles much faster)
rotate_extrude()
X(9)bentStrip([X(-3),Y(20),
turnXY(-45)*Y(10),
turnXY(-30)Y(15),
turnXY(60)(15),
turnXY(-30)*Y(20),X(8),Y(10)])
circle(d=2);
//as a rule of thumb, in the command List for bentStrip: if a command is
a turn put an asterisk after it, if it is a movement, like X() or Y()
put a comma. ony the last command does not need neither.
Peter