Re: is there a way to solid-fill an object?
If you want a solid filled object, conventional FDM 3D printing is the
wrong solution, solid filling does NOT make a component significantly
stronger, in directions other than along the layers [read grain for
timber] the walls and top/bottom do most of the work, if you are making
a structural component using FDM you need to either:
Reinforce the structure with strategically placed reinforcing systems
such as through bolts with large diameter washers on each end and a
nyloc nut.
or
Use a fibre reinforced material in the FDM process, it WILL still fail
in the layer direction first, but will be stiffer in the other directions.
or BOTH.
Account also needs to be made for creep, the process of deformation
under load over time.
To the issue of using a 3D scan:
If you are scanning with the intention of producing masters from which
to use 'lost wax' casting processes, you do not need a draft angle,
draft angles are only required when removing either a pattern from a
sand mould or for removing a casting/moulding, and I'm thinking here of
either injection moulding any component in a thermo plastic material, or
metal casting in a metal mould, usually, but not exclusively, zinc
alloys or aluminium.
The second and more important issue is that of shrinkage :-(
Metal/alloy at its melting point takes up more volume than at say room
temperature, allowance has to be made in the moulding process - i.e. the
master has to be larger than the final product to allow for the cooling
shrinkage as the component cools. Pattern makers of old used special
rulers, graduations in inches which were say 3% bigger than a 'standard
inch', the increase depended on the metal used for the casting.
Accordingly a 3D scan will not produce a model which will reproduce
at the correct, intended dimensions, but you can scale up a model in
Cura by a percentage in any direction.
For anybody interested, Prof J.E. Gordon's 'The New Science of Strong
Materials' [highly recommended] and Machinery's Handbook will fill in
the [necessarily] huge gaps in the above.
Apologies in advance for taking this way beyond the normal scope of
OpenSCAD, but we use OpenSCAD to solve real engineering problems and the
crossover from a model to a real object is sometimes a lot more
complicated - Gene's woodworking vice being a great example.
Roger,
[and I don't normally include this], B.Sc. Hons. Metallurgy and
Materials Science, 1976, University of Nottingham, England
I recently had a PLA part break so I sent the same STL to PCBway and got it
printed in 316L stainless steel by SLM for about $30. I don't think that
will ever break or creep!
[image: IMG_20250713_200539321.jpg]
[image: IMG_20250710_205902699.jpg]
It has the FDM techniques such as hanging holes, polyholes and teardrop
holes which aren't needed but don't cause a problem.
On Mon, 14 Jul 2025 at 10:39, Roger Whiteley via Discuss <
discuss@lists.openscad.org> wrote:
If you want a solid filled object, conventional FDM 3D printing is the
wrong solution, solid filling does NOT make a component significantly
stronger, in directions other than along the layers [read grain for
timber] the walls and top/bottom do most of the work, if you are making
a structural component using FDM you need to either:
Reinforce the structure with strategically placed reinforcing systems
such as through bolts with large diameter washers on each end and a
nyloc nut.
or
Use a fibre reinforced material in the FDM process, it WILL still fail
in the layer direction first, but will be stiffer in the other directions.
or BOTH.
Account also needs to be made for creep, the process of deformation
under load over time.
To the issue of using a 3D scan:
If you are scanning with the intention of producing masters from which
to use 'lost wax' casting processes, you do not need a draft angle,
draft angles are only required when removing either a pattern from a
sand mould or for removing a casting/moulding, and I'm thinking here of
either injection moulding any component in a thermo plastic material, or
metal casting in a metal mould, usually, but not exclusively, zinc
alloys or aluminium.
The second and more important issue is that of shrinkage :-(
Metal/alloy at its melting point takes up more volume than at say room
temperature, allowance has to be made in the moulding process - i.e. the
master has to be larger than the final product to allow for the cooling
shrinkage as the component cools. Pattern makers of old used special
rulers, graduations in inches which were say 3% bigger than a 'standard
inch', the increase depended on the metal used for the casting.
Accordingly a 3D scan will not produce a model which will reproduce
at the correct, intended dimensions, but you can scale up a model in
Cura by a percentage in any direction.
For anybody interested, Prof J.E. Gordon's 'The New Science of Strong
Materials' [highly recommended] and Machinery's Handbook will fill in
the [necessarily] huge gaps in the above.
Apologies in advance for taking this way beyond the normal scope of
OpenSCAD, but we use OpenSCAD to solve real engineering problems and the
crossover from a model to a real object is sometimes a lot more
complicated - Gene's woodworking vice being a great example.
Roger,
[and I don't normally include this], B.Sc. Hons. Metallurgy and
Materials Science, 1976, University of Nottingham, England
OpenSCAD mailing list
To unsubscribe send an email to discuss-leave@lists.openscad.org
Sigh... sorry, I have not found a suitable way to convey what I am
interested in doing. My apologies. I do not care about 100% infill in3D
printing, but creating an STL model of a part that has no overhangs.
Mathematically, I think this is called a 2D scaler field -- think of a GIS
and a map of the land. It can represent mountains and valleys, but not
caves (a cave, even when it forms a closed manifold surface, fold under
itself). I am looking for a tool that will modify a 3D model that has
caves in it, to one that replaces the cave mouth with a straight cliff. I
still do not feel I am explaining it, but I'll read through all the replies
and see if MeshLab, or other suggestions can be made to work.
Thanks again, all
On Mon, Jul 14, 2025 at 7:30 AM nop head via Discuss <
discuss@lists.openscad.org> wrote:
I recently had a PLA part break so I sent the same STL to PCBway and got
it printed in 316L stainless steel by SLM for about $30. I don't think
that will ever break or creep!
[image: IMG_20250713_200539321.jpg]
[image: IMG_20250710_205902699.jpg]
It has the FDM techniques such as hanging holes, polyholes and teardrop
holes which aren't needed but don't cause a problem.
On Mon, 14 Jul 2025 at 10:39, Roger Whiteley via Discuss <
discuss@lists.openscad.org> wrote:
If you want a solid filled object, conventional FDM 3D printing is the
wrong solution, solid filling does NOT make a component significantly
stronger, in directions other than along the layers [read grain for
timber] the walls and top/bottom do most of the work, if you are making
a structural component using FDM you need to either:
Reinforce the structure with strategically placed reinforcing systems
such as through bolts with large diameter washers on each end and a
nyloc nut.
or
Use a fibre reinforced material in the FDM process, it WILL still fail
in the layer direction first, but will be stiffer in the other directions.
or BOTH.
Account also needs to be made for creep, the process of deformation
under load over time.
To the issue of using a 3D scan:
If you are scanning with the intention of producing masters from which
to use 'lost wax' casting processes, you do not need a draft angle,
draft angles are only required when removing either a pattern from a
sand mould or for removing a casting/moulding, and I'm thinking here of
either injection moulding any component in a thermo plastic material, or
metal casting in a metal mould, usually, but not exclusively, zinc
alloys or aluminium.
The second and more important issue is that of shrinkage :-(
Metal/alloy at its melting point takes up more volume than at say room
temperature, allowance has to be made in the moulding process - i.e. the
master has to be larger than the final product to allow for the cooling
shrinkage as the component cools. Pattern makers of old used special
rulers, graduations in inches which were say 3% bigger than a 'standard
inch', the increase depended on the metal used for the casting.
Accordingly a 3D scan will not produce a model which will reproduce
at the correct, intended dimensions, but you can scale up a model in
Cura by a percentage in any direction.
For anybody interested, Prof J.E. Gordon's 'The New Science of Strong
Materials' [highly recommended] and Machinery's Handbook will fill in
the [necessarily] huge gaps in the above.
Apologies in advance for taking this way beyond the normal scope of
OpenSCAD, but we use OpenSCAD to solve real engineering problems and the
crossover from a model to a real object is sometimes a lot more
complicated - Gene's woodworking vice being a great example.
Roger,
[and I don't normally include this], B.Sc. Hons. Metallurgy and
Materials Science, 1976, University of Nottingham, England
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
On 7/14/2025 5:04 AM, John David via Discuss wrote:
Sigh... sorry, I have not found a suitable way to convey what I am
interested in doing. My apologies. I do not care about 100% infill
in3D printing, but creating an STL model of a part that has no
overhangs. Mathematically, I think this is called a 2D scaler field
-- think of a GIS and a map of the land. It can represent mountains
and valleys, but not caves (a cave, even when it forms a closed
manifold surface, fold under itself). I am looking for a tool that
will modify a 3D model that has caves in it, to one that replaces the
cave mouth with a straight cliff. I still do not feel I am explaining
it, but I'll read through all the replies and see if MeshLab, or other
suggestions can be made to work.
Hmm. Mountains and valleys but not caves is sometimes called 2.5D.
It's what a simple 3-axis mill can do.
Once you have a manifold object (which I think you didn't have at one
point), you could think of the problem as projecting a 3D object down
into 2.5D.
There's no simple way to do that with OpenSCAD, but you could
approximate it by unioning the object with copies of itself, stepped
down vertically, and then clipping at the bottom.
Here's a simple example:
module m() {
difference() {
cube(10);
translate([5,5,5]) rotate([90,0,0]) cylinder(h=11, d=3, center=true);
}
}
BIG = 1000;
module project25D(h, step) {
intersection() {
steps = floor(h/step);
for (i=[0:1:steps]) {
translate([0,0,-i*step])
children();
}
translate([-BIG/2, -BIG/2, 0]) cube(BIG);
}
}
project25D(10, 1) m();
But note: this won't address small features. I believe the limit is
that the step size must be smaller than any vertical feature. In
particular, if you have a pointy overhang, the result will be ragged:
project25D(10, 1) rotate([0,-45,0]) m();
becomes
On 7/14/25 07:35, nop head via Discuss wrote:
I recently had a PLA part break so I sent the same STL to PCBway and got it
printed in 316L stainless steel by SLM for about $30. I don't think that
will ever break or creep!
[image: IMG_20250713_200539321.jpg]
[image: IMG_20250710_205902699.jpg]
It has the FDM techniques such as hanging holes, polyholes and teardrop
holes which aren't needed but don't cause a problem.
On Mon, 14 Jul 2025 at 10:39, Roger Whiteley via Discuss <
discuss@lists.openscad.org> wrote:
If you want a solid filled object, conventional FDM 3D printing is the
wrong solution, solid filling does NOT make a component significantly
stronger, in directions other than along the layers [read grain for
timber] the walls and top/bottom do most of the work, if you are making
a structural component using FDM you need to either:
Reinforce the structure with strategically placed reinforcing systems
such as through bolts with large diameter washers on each end and a
nyloc nut.
or
Use a fibre reinforced material in the FDM process, it WILL still fail
in the layer direction first, but will be stiffer in the other directions.
or BOTH.
Account also needs to be made for creep, the process of deformation
under load over time.
To the issue of using a 3D scan:
If you are scanning with the intention of producing masters from which
to use 'lost wax' casting processes, you do not need a draft angle,
draft angles are only required when removing either a pattern from a
sand mould or for removing a casting/moulding, and I'm thinking here of
either injection moulding any component in a thermo plastic material, or
metal casting in a metal mould, usually, but not exclusively, zinc
alloys or aluminium.
The second and more important issue is that of shrinkage :-(
Metal/alloy at its melting point takes up more volume than at say room
temperature, allowance has to be made in the moulding process - i.e. the
master has to be larger than the final product to allow for the cooling
shrinkage as the component cools. Pattern makers of old used special
rulers, graduations in inches which were say 3% bigger than a 'standard
inch', the increase depended on the metal used for the casting.
Accordingly a 3D scan will not produce a model which will reproduce
at the correct, intended dimensions, but you can scale up a model in
Cura by a percentage in any direction.
For anybody interested, Prof J.E. Gordon's 'The New Science of Strong
Materials' [highly recommended] and Machinery's Handbook will fill in
the [necessarily] huge gaps in the above.
Apologies in advance for taking this way beyond the normal scope of
OpenSCAD, but we use OpenSCAD to solve real engineering problems and the
crossover from a model to a real object is sometimes a lot more
complicated - Gene's woodworking vice being a great example.
Roger,
I was not aware that a completed pix had ever been submitted to this
list. And someplace in the list settings, tbird isn't seeing the example
postings many others make to this list. So I am not seeing your posted
examples and haven't for about a year. I do recall posting a pix of the
printed half nuts though. So this pix is to try & see where its being
stripped by testing if I get it back.
This attached pix is of an earlier version, later ones I'm making are
about 4" longer, with a stronger internal design. The screw is nominally
2" in diameter but varies some as I quit the thread carving when the
flats of the original 2"x2" of hard maple disappear. The nuts, here
hidden inside the black PETG+CF have been lengthened to engage at least
12 of the carved threads. I figure on the resilience of PETG to
equalize the stress on the maple so it can exert a ton or more of
pressure on the vice jaw it will be incorporated into w/o ever damaging
the screw. I want the buyer to be able to use it the rest of his life
and hand the workbench its attached to to a grandchild interested in
working with wood for a living. There is not anything like it currently
available. I had hoped when I started this project, to get it announced
for sale, but diabetes keeps reminding me my time is limited since I'm
already well past my use by date at 90 yo.
So I'll see if I get this pix back. And if I don't get it back, somebody
can figure out why and fix it. Not being able to see your examples is a
bummer on a list such as this one is.
[and I don't normally include this], B.Sc. Hons. Metallurgy and
Materials Science, 1976, University of Nottingham, England
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
Cheers, Gene Heskett, CET.
"There are four boxes to be used in defense of liberty:
soap, ballot, jury, and ammo. Please use in that order."
-Ed Howdershelt (Author, 1940)
If we desire respect for the law, we must first make the law respectable.
- Louis D. Brandeis
On 7/14/2025 11:51 AM, gene heskett via Discuss wrote:
I was not aware that a completed pix had ever been submitted to this
list. And someplace in the list settings, tbird isn't seeing the
example postings many others make to this list. So I am not seeing
your posted examples and haven't for about a year. I do recall
posting a pix of the printed half nuts though. So this pix is to try &
see where its being stripped by testing if I get it back.
The problem is, I believe, that your Thunderbird is set to display plain
text, not HTML.
Strangely, I don't find the option in the hamburger menu (the three
lines in the top right). In the traditional menus, it's View / Message
Body As ... / Original HTML. You can get the traditional menus by
tapping the Alt key.
Thank you, @Jordan Brown openscad@jordan.maileater.net, your step/union
worked with minor tweaking. I had been looking to see if I could use sweep
or extrude, but as far as I could tell you can only sweep/extrude a 2D
object.
Anyway, since I have something that works, I'll call it a wrap. Thanks for
the solution.
EBo --
On Mon, Jul 14, 2025 at 3:03 PM Jordan Brown via Discuss <
discuss@lists.openscad.org> wrote:
On 7/14/2025 11:51 AM, gene heskett via Discuss wrote:
I was not aware that a completed pix had ever been submitted to this list.
And someplace in the list settings, tbird isn't seeing the example postings
many others make to this list. So I am not seeing your posted examples and
haven't for about a year. I do recall posting a pix of the printed half
nuts though. So this pix is to try & see where its being stripped by
testing if I get it back.
The problem is, I believe, that your Thunderbird is set to display plain
text, not HTML.
Strangely, I don't find the option in the hamburger menu (the three lines
in the top right). In the traditional menus, it's View / Message Body As
... / Original HTML. You can get the traditional menus by tapping the Alt
key.
OpenSCAD mailing list
To unsubscribe send an email to discuss-leave@lists.openscad.org
On 7/15/25 04:14, John David via Discuss wrote:
Thank you, @Jordan Brown openscad@jordan.maileater.net, your step/union
worked with minor tweaking. I had been looking to see if I could use sweep
or extrude, but as far as I could tell you can only sweep/extrude a 2D
object.
Anyway, since I have something that works, I'll call it a wrap. Thanks for
the solution.
EBo --
On Mon, Jul 14, 2025 at 3:03 PM Jordan Brown via Discuss <
discuss@lists.openscad.org> wrote:
On 7/14/2025 11:51 AM, gene heskett via Discuss wrote:
I was not aware that a completed pix had ever been submitted to this list.
And someplace in the list settings, tbird isn't seeing the example postings
many others make to this list. So I am not seeing your posted examples and
haven't for about a year. I do recall posting a pix of the printed half
nuts though. So this pix is to try & see where its being stripped by
testing if I get it back.
I am set for automatic, and I did not get the picture back, no mention
of the attachment in the return msg's.
The problem is, I believe, that your Thunderbird is set to display plain
text, not HTML.
Strangely, I don't find the option in the hamburger menu (the three lines
in the top right). In the traditional menus, it's View / Message Body As
... / Original HTML. You can get the traditional menus by tapping the Alt
key.
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
Cheers, Gene Heskett, CET.
"There are four boxes to be used in defense of liberty:
soap, ballot, jury, and ammo. Please use in that order."
-Ed Howdershelt (Author, 1940)
If we desire respect for the law, we must first make the law respectable.
- Louis D. Brandeis
It's bit ugly, but you CAN Minkowski the shape along an infinitesimally thin but tall cube.
-Revar
On Jul 15, 2025, at 3:23 AM, gene heskett via Discuss discuss@lists.openscad.org wrote:
On 7/15/25 04:14, John David via Discuss wrote:
Thank you, @Jordan Brown openscad@jordan.maileater.net, your step/union
worked with minor tweaking. I had been looking to see if I could use sweep
or extrude, but as far as I could tell you can only sweep/extrude a 2D
object.Anyway, since I have something that works, I'll call it a wrap. Thanks for
the solution.EBo --
On Mon, Jul 14, 2025 at 3:03 PM Jordan Brown via Discuss <
discuss@lists.openscad.org> wrote:On 7/14/2025 11:51 AM, gene heskett via Discuss wrote:
I was not aware that a completed pix had ever been submitted to this list.
And someplace in the list settings, tbird isn't seeing the example postings
many others make to this list. So I am not seeing your posted examples and
haven't for about a year. I do recall posting a pix of the printed half
nuts though. So this pix is to try & see where its being stripped by
testing if I get it back.
I am set for automatic, and I did not get the picture back, no mention of the attachment in the return msg's.The problem is, I believe, that your Thunderbird is set to display plain
text, not HTML.Strangely, I don't find the option in the hamburger menu (the three lines
in the top right). In the traditional menus, it's View / Message Body As
... / Original HTML. You can get the traditional menus by tapping the Alt
key.
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
Cheers, Gene Heskett, CET.
"There are four boxes to be used in defense of liberty:
soap, ballot, jury, and ammo. Please use in that order."
-Ed Howdershelt (Author, 1940)
If we desire respect for the law, we must first make the law respectable.
- Louis D. Brandeis
OpenSCAD mailing list
To unsubscribe send an email to discuss-leave@lists.openscad.org