Thanks for the explanation Can you share a hand sketch of what will be the final expected figure. Maybe someone here has already written something similar. On Sun, 16 Oct, 2022, 12:55 am neri-engineering via Discuss, < discuss@lists.openscad.org> wrote: > > > > ---------- Forwarded message ---------- > From: neri-engineering <neri-engineering@protonmail.com> > To: OpenSCAD general discussion Mailing-list <discuss@lists.openscad.org> > Cc: > Bcc: > Date: Sat, 15 Oct 2022 19:24:32 +0000 > Subject: [OpenSCAD] Re: multiple section extrude > Long ago I wrote something similar, in order to generate a Klein bottle, > and for that reason I can offer a suggestion. > > The "path of center" could be a smooth curve, e.g. Bezier or cubic curve, > where a discrete set of points defines the actual path (center curve passes > through points) or the points serve as "control points". I forget the > technical name for these sorts of curve, I only remember that there are > several variations, and I think they involved some sort of cubic curves. > > The radius as it varies as you travel through the center curve could also > be a smooth curve, following a similar strategy. Control points or actual > radii at sample points could determine the radius along every point of the > center curve. > > So in essence what I'm proposing are two "smooth" functions in terms of > 't' which is in the interval [0,1]. > > First function C:[0,1] -> R3 maps every point in the interval [0,1] to a > point in 3-space and the smooth curve defined such is the center of your > "tube". > > Second function D:[0,1] -> R maps every point in the interval [0,1] to a > real value, which is defined to be the radius of tube at that location. > > Of course you'd have to do things like figuring out how to trace the > circle at every point, based on some simple vector calculations, to > determine perpendiculars etc. Simple vector calculations, dot products, > cross products, and the like. > > You'd get a very smooth tube which would be highly desirable. > > I'm only sharing this because I had great success with the idea. > > https://odysee.com/@Neri_Engineering/cyclist-on-klein-bottle > > Sent with Proton Mail <https://proton.me/> secure email. > > ------- Original Message ------- > On Saturday, October 15th, 2022 at 12:04 PM, Sanjeev Prabhakar < > sprabhakar2006@gmail.com> wrote: > > I wrote a function "multiple section extrude" > > Idea is supply path points, sections for each path points and radiuses for > each path point. This will create a polyhedron with multi_sections like the > example picture shown below. > > I am sure the similar function can be written in original openscad > > example: > > https://github.com/sprabhakar2006/openSCAD/blob/main/multiple_section_extrude.ipynb > > > https://github.com/sprabhakar2006/openSCAD/blob/main/multiple_section_extrude.scad > [image: Screenshot 2022-10-15 at 10.14.37 PM.png] > > > > > > ---------- Forwarded message ---------- > From: neri-engineering via Discuss <discuss@lists.openscad.org> > To: OpenSCAD general discussion Mailing-list <discuss@lists.openscad.org> > Cc: neri-engineering <neri-engineering@protonmail.com> > Bcc: > Date: Sat, 15 Oct 2022 19:24:32 +0000 > Subject: [OpenSCAD] Re: multiple section extrude > _______________________________________________ > OpenSCAD mailing list > To unsubscribe send an email to discuss-leave@lists.openscad.org >

neri-engineering is just talking about a basic sweep algorithm applied to a circular cross section, which length-dependent scale. It's what the original poster was donig with the neck1200 model. He was suggesting creating the path with beziers or some other method to produce a smooth path. Since he assumes a circular cross section, all the complications having to do with twisting disappear. On Sat, Oct 15, 2022 at 8:36 PM Sanjeev Prabhakar <sprabhakar2006@gmail.com> wrote: > Thanks for the explanation > Can you share a hand sketch of what will be the final expected figure. > Maybe someone here has already written something similar. > > On Sun, 16 Oct, 2022, 12:55 am neri-engineering via Discuss, < > discuss@lists.openscad.org> wrote: > >> >> >> >> ---------- Forwarded message ---------- >> From: neri-engineering <neri-engineering@protonmail.com> >> To: OpenSCAD general discussion Mailing-list <discuss@lists.openscad.org> >> Cc: >> Bcc: >> Date: Sat, 15 Oct 2022 19:24:32 +0000 >> Subject: [OpenSCAD] Re: multiple section extrude >> Long ago I wrote something similar, in order to generate a Klein bottle, >> and for that reason I can offer a suggestion. >> >> The "path of center" could be a smooth curve, e.g. Bezier or cubic curve, >> where a discrete set of points defines the actual path (center curve passes >> through points) or the points serve as "control points". I forget the >> technical name for these sorts of curve, I only remember that there are >> several variations, and I think they involved some sort of cubic curves. >> >> The radius as it varies as you travel through the center curve could also >> be a smooth curve, following a similar strategy. Control points or actual >> radii at sample points could determine the radius along every point of the >> center curve. >> >> So in essence what I'm proposing are two "smooth" functions in terms of >> 't' which is in the interval [0,1]. >> >> First function C:[0,1] -> R3 maps every point in the interval [0,1] to a >> point in 3-space and the smooth curve defined such is the center of your >> "tube". >> >> Second function D:[0,1] -> R maps every point in the interval [0,1] to a >> real value, which is defined to be the radius of tube at that location. >> >> Of course you'd have to do things like figuring out how to trace the >> circle at every point, based on some simple vector calculations, to >> determine perpendiculars etc. Simple vector calculations, dot products, >> cross products, and the like. >> >> You'd get a very smooth tube which would be highly desirable. >> >> I'm only sharing this because I had great success with the idea. >> >> https://odysee.com/@Neri_Engineering/cyclist-on-klein-bottle >> >> Sent with Proton Mail <https://proton.me/> secure email. >> >> ------- Original Message ------- >> On Saturday, October 15th, 2022 at 12:04 PM, Sanjeev Prabhakar < >> sprabhakar2006@gmail.com> wrote: >> >> I wrote a function "multiple section extrude" >> >> Idea is supply path points, sections for each path points and radiuses >> for each path point. This will create a polyhedron with multi_sections like >> the example picture shown below. >> >> I am sure the similar function can be written in original openscad >> >> example: >> >> https://github.com/sprabhakar2006/openSCAD/blob/main/multiple_section_extrude.ipynb >> >> >> https://github.com/sprabhakar2006/openSCAD/blob/main/multiple_section_extrude.scad >> [image: Screenshot 2022-10-15 at 10.14.37 PM.png] >> >> >> >> >> >> ---------- Forwarded message ---------- >> From: neri-engineering via Discuss <discuss@lists.openscad.org> >> To: OpenSCAD general discussion Mailing-list <discuss@lists.openscad.org> >> Cc: neri-engineering <neri-engineering@protonmail.com> >> Bcc: >> Date: Sat, 15 Oct 2022 19:24:32 +0000 >> Subject: [OpenSCAD] Re: multiple section extrude >> _______________________________________________ >> 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 >

OK, if that is the case then it is a bit simpler On Sun, 16 Oct 2022 at 06:12, Adrian Mariano <avm4@cornell.edu> wrote: > neri-engineering is just talking about a basic sweep algorithm applied to > a circular cross section, which length-dependent scale. It's what the > original poster was donig with the neck1200 model. He was suggesting > creating the path with beziers or some other method to produce a smooth > path. Since he assumes a circular cross section, all the complications > having to do with twisting disappear. > > On Sat, Oct 15, 2022 at 8:36 PM Sanjeev Prabhakar < > sprabhakar2006@gmail.com> wrote: > >> Thanks for the explanation >> Can you share a hand sketch of what will be the final expected figure. >> Maybe someone here has already written something similar. >> >> On Sun, 16 Oct, 2022, 12:55 am neri-engineering via Discuss, < >> discuss@lists.openscad.org> wrote: >> >>> >>> >>> >>> ---------- Forwarded message ---------- >>> From: neri-engineering <neri-engineering@protonmail.com> >>> To: OpenSCAD general discussion Mailing-list <discuss@lists.openscad.org >>> > >>> Cc: >>> Bcc: >>> Date: Sat, 15 Oct 2022 19:24:32 +0000 >>> Subject: [OpenSCAD] Re: multiple section extrude >>> Long ago I wrote something similar, in order to generate a Klein bottle, >>> and for that reason I can offer a suggestion. >>> >>> The "path of center" could be a smooth curve, e.g. Bezier or cubic >>> curve, where a discrete set of points defines the actual path (center curve >>> passes through points) or the points serve as "control points". I forget >>> the technical name for these sorts of curve, I only remember that there are >>> several variations, and I think they involved some sort of cubic curves. >>> >>> The radius as it varies as you travel through the center curve could >>> also be a smooth curve, following a similar strategy. Control points or >>> actual radii at sample points could determine the radius along every point >>> of the center curve. >>> >>> So in essence what I'm proposing are two "smooth" functions in terms of >>> 't' which is in the interval [0,1]. >>> >>> First function C:[0,1] -> R3 maps every point in the interval [0,1] to a >>> point in 3-space and the smooth curve defined such is the center of your >>> "tube". >>> >>> Second function D:[0,1] -> R maps every point in the interval [0,1] to a >>> real value, which is defined to be the radius of tube at that location. >>> >>> Of course you'd have to do things like figuring out how to trace the >>> circle at every point, based on some simple vector calculations, to >>> determine perpendiculars etc. Simple vector calculations, dot products, >>> cross products, and the like. >>> >>> You'd get a very smooth tube which would be highly desirable. >>> >>> I'm only sharing this because I had great success with the idea. >>> >>> https://odysee.com/@Neri_Engineering/cyclist-on-klein-bottle >>> >>> Sent with Proton Mail <https://proton.me/> secure email. >>> >>> ------- Original Message ------- >>> On Saturday, October 15th, 2022 at 12:04 PM, Sanjeev Prabhakar < >>> sprabhakar2006@gmail.com> wrote: >>> >>> I wrote a function "multiple section extrude" >>> >>> Idea is supply path points, sections for each path points and radiuses >>> for each path point. This will create a polyhedron with multi_sections like >>> the example picture shown below. >>> >>> I am sure the similar function can be written in original openscad >>> >>> example: >>> >>> https://github.com/sprabhakar2006/openSCAD/blob/main/multiple_section_extrude.ipynb >>> >>> >>> https://github.com/sprabhakar2006/openSCAD/blob/main/multiple_section_extrude.scad >>> [image: Screenshot 2022-10-15 at 10.14.37 PM.png] >>> >>> >>> >>> >>> >>> ---------- Forwarded message ---------- >>> From: neri-engineering via Discuss <discuss@lists.openscad.org> >>> To: OpenSCAD general discussion Mailing-list <discuss@lists.openscad.org >>> > >>> Cc: neri-engineering <neri-engineering@protonmail.com> >>> Bcc: >>> Date: Sat, 15 Oct 2022 19:24:32 +0000 >>> Subject: [OpenSCAD] Re: multiple section extrude >>> _______________________________________________ >>> 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 >

another example of multiple section extrude function. It might seem trivial, but it is not. https://github.com/sprabhakar2006/openSCAD/blob/main/multiple_sec_extrude_example2.scad https://github.com/sprabhakar2006/openSCAD/blob/main/multiple_section_extrude.ipynb [image: Screenshot 2022-10-16 at 9.35.54 PM.png] On Sun, 16 Oct 2022 at 06:45, Sanjeev Prabhakar <sprabhakar2006@gmail.com> wrote: > OK, if that is the case then it is a bit simpler > > On Sun, 16 Oct 2022 at 06:12, Adrian Mariano <avm4@cornell.edu> wrote: > >> neri-engineering is just talking about a basic sweep algorithm applied to >> a circular cross section, which length-dependent scale. It's what the >> original poster was donig with the neck1200 model. He was suggesting >> creating the path with beziers or some other method to produce a smooth >> path. Since he assumes a circular cross section, all the complications >> having to do with twisting disappear. >> >> On Sat, Oct 15, 2022 at 8:36 PM Sanjeev Prabhakar < >> sprabhakar2006@gmail.com> wrote: >> >>> Thanks for the explanation >>> Can you share a hand sketch of what will be the final expected figure. >>> Maybe someone here has already written something similar. >>> >>> On Sun, 16 Oct, 2022, 12:55 am neri-engineering via Discuss, < >>> discuss@lists.openscad.org> wrote: >>> >>>> >>>> >>>> >>>> ---------- Forwarded message ---------- >>>> From: neri-engineering <neri-engineering@protonmail.com> >>>> To: OpenSCAD general discussion Mailing-list < >>>> discuss@lists.openscad.org> >>>> Cc: >>>> Bcc: >>>> Date: Sat, 15 Oct 2022 19:24:32 +0000 >>>> Subject: [OpenSCAD] Re: multiple section extrude >>>> Long ago I wrote something similar, in order to generate a Klein >>>> bottle, and for that reason I can offer a suggestion. >>>> >>>> The "path of center" could be a smooth curve, e.g. Bezier or cubic >>>> curve, where a discrete set of points defines the actual path (center curve >>>> passes through points) or the points serve as "control points". I forget >>>> the technical name for these sorts of curve, I only remember that there are >>>> several variations, and I think they involved some sort of cubic curves. >>>> >>>> The radius as it varies as you travel through the center curve could >>>> also be a smooth curve, following a similar strategy. Control points or >>>> actual radii at sample points could determine the radius along every point >>>> of the center curve. >>>> >>>> So in essence what I'm proposing are two "smooth" functions in terms of >>>> 't' which is in the interval [0,1]. >>>> >>>> First function C:[0,1] -> R3 maps every point in the interval [0,1] to >>>> a point in 3-space and the smooth curve defined such is the center of your >>>> "tube". >>>> >>>> Second function D:[0,1] -> R maps every point in the interval [0,1] to >>>> a real value, which is defined to be the radius of tube at that location. >>>> >>>> Of course you'd have to do things like figuring out how to trace the >>>> circle at every point, based on some simple vector calculations, to >>>> determine perpendiculars etc. Simple vector calculations, dot products, >>>> cross products, and the like. >>>> >>>> You'd get a very smooth tube which would be highly desirable. >>>> >>>> I'm only sharing this because I had great success with the idea. >>>> >>>> https://odysee.com/@Neri_Engineering/cyclist-on-klein-bottle >>>> >>>> Sent with Proton Mail <https://proton.me/> secure email. >>>> >>>> ------- Original Message ------- >>>> On Saturday, October 15th, 2022 at 12:04 PM, Sanjeev Prabhakar < >>>> sprabhakar2006@gmail.com> wrote: >>>> >>>> I wrote a function "multiple section extrude" >>>> >>>> Idea is supply path points, sections for each path points and radiuses >>>> for each path point. This will create a polyhedron with multi_sections like >>>> the example picture shown below. >>>> >>>> I am sure the similar function can be written in original openscad >>>> >>>> example: >>>> >>>> https://github.com/sprabhakar2006/openSCAD/blob/main/multiple_section_extrude.ipynb >>>> >>>> >>>> https://github.com/sprabhakar2006/openSCAD/blob/main/multiple_section_extrude.scad >>>> [image: Screenshot 2022-10-15 at 10.14.37 PM.png] >>>> >>>> >>>> >>>> >>>> >>>> ---------- Forwarded message ---------- >>>> From: neri-engineering via Discuss <discuss@lists.openscad.org> >>>> To: OpenSCAD general discussion Mailing-list < >>>> discuss@lists.openscad.org> >>>> Cc: neri-engineering <neri-engineering@protonmail.com> >>>> Bcc: >>>> Date: Sat, 15 Oct 2022 19:24:32 +0000 >>>> Subject: [OpenSCAD] Re: multiple section extrude >>>> _______________________________________________ >>>> 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 >> >

The section changes scale in one axis, but otherwise remains the same, no? On Sun, Oct 16, 2022 at 12:16 PM Sanjeev Prabhakar <sprabhakar2006@gmail.com> wrote: > another example of multiple section extrude function. > It might seem trivial, but it is not. > > > https://github.com/sprabhakar2006/openSCAD/blob/main/multiple_sec_extrude_example2.scad > > > https://github.com/sprabhakar2006/openSCAD/blob/main/multiple_section_extrude.ipynb > [image: Screenshot 2022-10-16 at 9.35.54 PM.png] > > On Sun, 16 Oct 2022 at 06:45, Sanjeev Prabhakar <sprabhakar2006@gmail.com> > wrote: > >> OK, if that is the case then it is a bit simpler >> >> On Sun, 16 Oct 2022 at 06:12, Adrian Mariano <avm4@cornell.edu> wrote: >> >>> neri-engineering is just talking about a basic sweep algorithm applied >>> to a circular cross section, which length-dependent scale. It's what the >>> original poster was donig with the neck1200 model. He was suggesting >>> creating the path with beziers or some other method to produce a smooth >>> path. Since he assumes a circular cross section, all the complications >>> having to do with twisting disappear. >>> >>> On Sat, Oct 15, 2022 at 8:36 PM Sanjeev Prabhakar < >>> sprabhakar2006@gmail.com> wrote: >>> >>>> Thanks for the explanation >>>> Can you share a hand sketch of what will be the final expected figure. >>>> Maybe someone here has already written something similar. >>>> >>>> On Sun, 16 Oct, 2022, 12:55 am neri-engineering via Discuss, < >>>> discuss@lists.openscad.org> wrote: >>>> >>>>> >>>>> >>>>> >>>>> ---------- Forwarded message ---------- >>>>> From: neri-engineering <neri-engineering@protonmail.com> >>>>> To: OpenSCAD general discussion Mailing-list < >>>>> discuss@lists.openscad.org> >>>>> Cc: >>>>> Bcc: >>>>> Date: Sat, 15 Oct 2022 19:24:32 +0000 >>>>> Subject: [OpenSCAD] Re: multiple section extrude >>>>> Long ago I wrote something similar, in order to generate a Klein >>>>> bottle, and for that reason I can offer a suggestion. >>>>> >>>>> The "path of center" could be a smooth curve, e.g. Bezier or cubic >>>>> curve, where a discrete set of points defines the actual path (center curve >>>>> passes through points) or the points serve as "control points". I forget >>>>> the technical name for these sorts of curve, I only remember that there are >>>>> several variations, and I think they involved some sort of cubic curves. >>>>> >>>>> The radius as it varies as you travel through the center curve could >>>>> also be a smooth curve, following a similar strategy. Control points or >>>>> actual radii at sample points could determine the radius along every point >>>>> of the center curve. >>>>> >>>>> So in essence what I'm proposing are two "smooth" functions in terms >>>>> of 't' which is in the interval [0,1]. >>>>> >>>>> First function C:[0,1] -> R3 maps every point in the interval [0,1] to >>>>> a point in 3-space and the smooth curve defined such is the center of your >>>>> "tube". >>>>> >>>>> Second function D:[0,1] -> R maps every point in the interval [0,1] to >>>>> a real value, which is defined to be the radius of tube at that location. >>>>> >>>>> Of course you'd have to do things like figuring out how to trace the >>>>> circle at every point, based on some simple vector calculations, to >>>>> determine perpendiculars etc. Simple vector calculations, dot products, >>>>> cross products, and the like. >>>>> >>>>> You'd get a very smooth tube which would be highly desirable. >>>>> >>>>> I'm only sharing this because I had great success with the idea. >>>>> >>>>> https://odysee.com/@Neri_Engineering/cyclist-on-klein-bottle >>>>> >>>>> Sent with Proton Mail <https://proton.me/> secure email. >>>>> >>>>> ------- Original Message ------- >>>>> On Saturday, October 15th, 2022 at 12:04 PM, Sanjeev Prabhakar < >>>>> sprabhakar2006@gmail.com> wrote: >>>>> >>>>> I wrote a function "multiple section extrude" >>>>> >>>>> Idea is supply path points, sections for each path points and radiuses >>>>> for each path point. This will create a polyhedron with multi_sections like >>>>> the example picture shown below. >>>>> >>>>> I am sure the similar function can be written in original openscad >>>>> >>>>> example: >>>>> >>>>> https://github.com/sprabhakar2006/openSCAD/blob/main/multiple_section_extrude.ipynb >>>>> >>>>> >>>>> https://github.com/sprabhakar2006/openSCAD/blob/main/multiple_section_extrude.scad >>>>> [image: Screenshot 2022-10-15 at 10.14.37 PM.png] >>>>> >>>>> >>>>> >>>>> >>>>> >>>>> ---------- Forwarded message ---------- >>>>> From: neri-engineering via Discuss <discuss@lists.openscad.org> >>>>> To: OpenSCAD general discussion Mailing-list < >>>>> discuss@lists.openscad.org> >>>>> Cc: neri-engineering <neri-engineering@protonmail.com> >>>>> Bcc: >>>>> Date: Sat, 15 Oct 2022 19:24:32 +0000 >>>>> Subject: [OpenSCAD] Re: multiple section extrude >>>>> _______________________________________________ >>>>> 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 >

No it's not scale There are 2 3d sections Both the sections have same number of points If you see the sections from top view both are rectangular, 1 larger and 1 smaller If you view the sections from " -y" the larger section will be almost a straight line and smaller section will be an arc (arc pointing on bottom side). I hope this could explain to some extent. You can read the code for better understanding. On Sun, 16 Oct, 2022, 10:02 pm Adrian Mariano, <avm4@cornell.edu> wrote: > The section changes scale in one axis, but otherwise remains the same, > no? > > On Sun, Oct 16, 2022 at 12:16 PM Sanjeev Prabhakar < > sprabhakar2006@gmail.com> wrote: > >> another example of multiple section extrude function. >> It might seem trivial, but it is not. >> >> >> https://github.com/sprabhakar2006/openSCAD/blob/main/multiple_sec_extrude_example2.scad >> >> >> https://github.com/sprabhakar2006/openSCAD/blob/main/multiple_section_extrude.ipynb >> [image: Screenshot 2022-10-16 at 9.35.54 PM.png] >> >> On Sun, 16 Oct 2022 at 06:45, Sanjeev Prabhakar <sprabhakar2006@gmail.com> >> wrote: >> >>> OK, if that is the case then it is a bit simpler >>> >>> On Sun, 16 Oct 2022 at 06:12, Adrian Mariano <avm4@cornell.edu> wrote: >>> >>>> neri-engineering is just talking about a basic sweep algorithm applied >>>> to a circular cross section, which length-dependent scale. It's what the >>>> original poster was donig with the neck1200 model. He was suggesting >>>> creating the path with beziers or some other method to produce a smooth >>>> path. Since he assumes a circular cross section, all the complications >>>> having to do with twisting disappear. >>>> >>>> On Sat, Oct 15, 2022 at 8:36 PM Sanjeev Prabhakar < >>>> sprabhakar2006@gmail.com> wrote: >>>> >>>>> Thanks for the explanation >>>>> Can you share a hand sketch of what will be the final expected figure. >>>>> Maybe someone here has already written something similar. >>>>> >>>>> On Sun, 16 Oct, 2022, 12:55 am neri-engineering via Discuss, < >>>>> discuss@lists.openscad.org> wrote: >>>>> >>>>>> >>>>>> >>>>>> >>>>>> ---------- Forwarded message ---------- >>>>>> From: neri-engineering <neri-engineering@protonmail.com> >>>>>> To: OpenSCAD general discussion Mailing-list < >>>>>> discuss@lists.openscad.org> >>>>>> Cc: >>>>>> Bcc: >>>>>> Date: Sat, 15 Oct 2022 19:24:32 +0000 >>>>>> Subject: [OpenSCAD] Re: multiple section extrude >>>>>> Long ago I wrote something similar, in order to generate a Klein >>>>>> bottle, and for that reason I can offer a suggestion. >>>>>> >>>>>> The "path of center" could be a smooth curve, e.g. Bezier or cubic >>>>>> curve, where a discrete set of points defines the actual path (center curve >>>>>> passes through points) or the points serve as "control points". I forget >>>>>> the technical name for these sorts of curve, I only remember that there are >>>>>> several variations, and I think they involved some sort of cubic curves. >>>>>> >>>>>> The radius as it varies as you travel through the center curve could >>>>>> also be a smooth curve, following a similar strategy. Control points or >>>>>> actual radii at sample points could determine the radius along every point >>>>>> of the center curve. >>>>>> >>>>>> So in essence what I'm proposing are two "smooth" functions in terms >>>>>> of 't' which is in the interval [0,1]. >>>>>> >>>>>> First function C:[0,1] -> R3 maps every point in the interval [0,1] >>>>>> to a point in 3-space and the smooth curve defined such is the center of >>>>>> your "tube". >>>>>> >>>>>> Second function D:[0,1] -> R maps every point in the interval [0,1] >>>>>> to a real value, which is defined to be the radius of tube at that location. >>>>>> >>>>>> Of course you'd have to do things like figuring out how to trace the >>>>>> circle at every point, based on some simple vector calculations, to >>>>>> determine perpendiculars etc. Simple vector calculations, dot products, >>>>>> cross products, and the like. >>>>>> >>>>>> You'd get a very smooth tube which would be highly desirable. >>>>>> >>>>>> I'm only sharing this because I had great success with the idea. >>>>>> >>>>>> https://odysee.com/@Neri_Engineering/cyclist-on-klein-bottle >>>>>> >>>>>> Sent with Proton Mail <https://proton.me/> secure email. >>>>>> >>>>>> ------- Original Message ------- >>>>>> On Saturday, October 15th, 2022 at 12:04 PM, Sanjeev Prabhakar < >>>>>> sprabhakar2006@gmail.com> wrote: >>>>>> >>>>>> I wrote a function "multiple section extrude" >>>>>> >>>>>> Idea is supply path points, sections for each path points and >>>>>> radiuses for each path point. This will create a polyhedron with >>>>>> multi_sections like the example picture shown below. >>>>>> >>>>>> I am sure the similar function can be written in original openscad >>>>>> >>>>>> example: >>>>>> >>>>>> https://github.com/sprabhakar2006/openSCAD/blob/main/multiple_section_extrude.ipynb >>>>>> >>>>>> >>>>>> https://github.com/sprabhakar2006/openSCAD/blob/main/multiple_section_extrude.scad >>>>>> [image: Screenshot 2022-10-15 at 10.14.37 PM.png] >>>>>> >>>>>> >>>>>> >>>>>> >>>>>> >>>>>> ---------- Forwarded message ---------- >>>>>> From: neri-engineering via Discuss <discuss@lists.openscad.org> >>>>>> To: OpenSCAD general discussion Mailing-list < >>>>>> discuss@lists.openscad.org> >>>>>> Cc: neri-engineering <neri-engineering@protonmail.com> >>>>>> Bcc: >>>>>> Date: Sat, 15 Oct 2022 19:24:32 +0000 >>>>>> Subject: [OpenSCAD] Re: multiple section extrude >>>>>> _______________________________________________ >>>>>> 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 >

for better understanding, see the picture [image: Screenshot 2022-10-17 at 6.39.01 AM.png] On Sun, 16 Oct 2022 at 22:41, Sanjeev Prabhakar <sprabhakar2006@gmail.com> wrote: > No it's not scale > > There are 2 3d sections > > Both the sections have same number of points > > If you see the sections from top view both are rectangular, 1 larger and > 1 smaller > > If you view the sections from " -y" the larger section will be almost a > straight line and smaller section will be an arc (arc pointing on bottom > side). > > I hope this could explain to some extent. > > You can read the code for better understanding. > > > > > > On Sun, 16 Oct, 2022, 10:02 pm Adrian Mariano, <avm4@cornell.edu> wrote: > >> The section changes scale in one axis, but otherwise remains the same, >> no? >> >> On Sun, Oct 16, 2022 at 12:16 PM Sanjeev Prabhakar < >> sprabhakar2006@gmail.com> wrote: >> >>> another example of multiple section extrude function. >>> It might seem trivial, but it is not. >>> >>> >>> https://github.com/sprabhakar2006/openSCAD/blob/main/multiple_sec_extrude_example2.scad >>> >>> >>> https://github.com/sprabhakar2006/openSCAD/blob/main/multiple_section_extrude.ipynb >>> [image: Screenshot 2022-10-16 at 9.35.54 PM.png] >>> >>> On Sun, 16 Oct 2022 at 06:45, Sanjeev Prabhakar < >>> sprabhakar2006@gmail.com> wrote: >>> >>>> OK, if that is the case then it is a bit simpler >>>> >>>> On Sun, 16 Oct 2022 at 06:12, Adrian Mariano <avm4@cornell.edu> wrote: >>>> >>>>> neri-engineering is just talking about a basic sweep algorithm applied >>>>> to a circular cross section, which length-dependent scale. It's what the >>>>> original poster was donig with the neck1200 model. He was suggesting >>>>> creating the path with beziers or some other method to produce a smooth >>>>> path. Since he assumes a circular cross section, all the complications >>>>> having to do with twisting disappear. >>>>> >>>>> On Sat, Oct 15, 2022 at 8:36 PM Sanjeev Prabhakar < >>>>> sprabhakar2006@gmail.com> wrote: >>>>> >>>>>> Thanks for the explanation >>>>>> Can you share a hand sketch of what will be the final expected figure. >>>>>> Maybe someone here has already written something similar. >>>>>> >>>>>> On Sun, 16 Oct, 2022, 12:55 am neri-engineering via Discuss, < >>>>>> discuss@lists.openscad.org> wrote: >>>>>> >>>>>>> >>>>>>> >>>>>>> >>>>>>> ---------- Forwarded message ---------- >>>>>>> From: neri-engineering <neri-engineering@protonmail.com> >>>>>>> To: OpenSCAD general discussion Mailing-list < >>>>>>> discuss@lists.openscad.org> >>>>>>> Cc: >>>>>>> Bcc: >>>>>>> Date: Sat, 15 Oct 2022 19:24:32 +0000 >>>>>>> Subject: [OpenSCAD] Re: multiple section extrude >>>>>>> Long ago I wrote something similar, in order to generate a Klein >>>>>>> bottle, and for that reason I can offer a suggestion. >>>>>>> >>>>>>> The "path of center" could be a smooth curve, e.g. Bezier or cubic >>>>>>> curve, where a discrete set of points defines the actual path (center curve >>>>>>> passes through points) or the points serve as "control points". I forget >>>>>>> the technical name for these sorts of curve, I only remember that there are >>>>>>> several variations, and I think they involved some sort of cubic curves. >>>>>>> >>>>>>> The radius as it varies as you travel through the center curve could >>>>>>> also be a smooth curve, following a similar strategy. Control points or >>>>>>> actual radii at sample points could determine the radius along every point >>>>>>> of the center curve. >>>>>>> >>>>>>> So in essence what I'm proposing are two "smooth" functions in terms >>>>>>> of 't' which is in the interval [0,1]. >>>>>>> >>>>>>> First function C:[0,1] -> R3 maps every point in the interval [0,1] >>>>>>> to a point in 3-space and the smooth curve defined such is the center of >>>>>>> your "tube". >>>>>>> >>>>>>> Second function D:[0,1] -> R maps every point in the interval [0,1] >>>>>>> to a real value, which is defined to be the radius of tube at that location. >>>>>>> >>>>>>> Of course you'd have to do things like figuring out how to trace the >>>>>>> circle at every point, based on some simple vector calculations, to >>>>>>> determine perpendiculars etc. Simple vector calculations, dot products, >>>>>>> cross products, and the like. >>>>>>> >>>>>>> You'd get a very smooth tube which would be highly desirable. >>>>>>> >>>>>>> I'm only sharing this because I had great success with the idea. >>>>>>> >>>>>>> https://odysee.com/@Neri_Engineering/cyclist-on-klein-bottle >>>>>>> >>>>>>> Sent with Proton Mail <https://proton.me/> secure email. >>>>>>> >>>>>>> ------- Original Message ------- >>>>>>> On Saturday, October 15th, 2022 at 12:04 PM, Sanjeev Prabhakar < >>>>>>> sprabhakar2006@gmail.com> wrote: >>>>>>> >>>>>>> I wrote a function "multiple section extrude" >>>>>>> >>>>>>> Idea is supply path points, sections for each path points and >>>>>>> radiuses for each path point. This will create a polyhedron with >>>>>>> multi_sections like the example picture shown below. >>>>>>> >>>>>>> I am sure the similar function can be written in original openscad >>>>>>> >>>>>>> example: >>>>>>> >>>>>>> https://github.com/sprabhakar2006/openSCAD/blob/main/multiple_section_extrude.ipynb >>>>>>> >>>>>>> >>>>>>> https://github.com/sprabhakar2006/openSCAD/blob/main/multiple_section_extrude.scad >>>>>>> [image: Screenshot 2022-10-15 at 10.14.37 PM.png] >>>>>>> >>>>>>> >>>>>>> >>>>>>> >>>>>>> >>>>>>> ---------- Forwarded message ---------- >>>>>>> From: neri-engineering via Discuss <discuss@lists.openscad.org> >>>>>>> To: OpenSCAD general discussion Mailing-list < >>>>>>> discuss@lists.openscad.org> >>>>>>> Cc: neri-engineering <neri-engineering@protonmail.com> >>>>>>> Bcc: >>>>>>> Date: Sat, 15 Oct 2022 19:24:32 +0000 >>>>>>> Subject: [OpenSCAD] Re: multiple section extrude >>>>>>> _______________________________________________ >>>>>>> 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 think I understand, but I had a different approach for how to make the shape. You don't need to build this from the curved section 2 that you show, because you can create it as a series of slices of a rectangle: a scaled path_sweep. I tried to roughly duplicate the shape using this idea in BOSL2. (Note: this is unreleased code so it won't run yet on the posted BOSL2, which doesn't support scaling.) opath = [[0,0,0], [100,0,0], [200, 0, 33]]; path=subdivide_path(opath, 100,closed=false); scales = [each repeat([1,1],50), for(i=[1:-1/50:1/50]) [sqrt(i),1] ]; path_sweep(rect([150,10]),path,scale=scales); The point is that the shape is just a rectangle scaled in its width with a continuously decreasing width that approaches zero along a curved path for the end. [image: image.png] On Sun, Oct 16, 2022 at 9:16 PM Sanjeev Prabhakar <sprabhakar2006@gmail.com> wrote: > for better understanding, see the picture > [image: Screenshot 2022-10-17 at 6.39.01 AM.png] > > On Sun, 16 Oct 2022 at 22:41, Sanjeev Prabhakar <sprabhakar2006@gmail.com> > wrote: > >> No it's not scale >> >> There are 2 3d sections >> >> Both the sections have same number of points >> >> If you see the sections from top view both are rectangular, 1 larger and >> 1 smaller >> >> If you view the sections from " -y" the larger section will be almost a >> straight line and smaller section will be an arc (arc pointing on bottom >> side). >> >> I hope this could explain to some extent. >> >> You can read the code for better understanding. >> >> >> >> >> >> On Sun, 16 Oct, 2022, 10:02 pm Adrian Mariano, <avm4@cornell.edu> wrote: >> >>> The section changes scale in one axis, but otherwise remains the same, >>> no? >>> >>> On Sun, Oct 16, 2022 at 12:16 PM Sanjeev Prabhakar < >>> sprabhakar2006@gmail.com> wrote: >>> >>>> another example of multiple section extrude function. >>>> It might seem trivial, but it is not. >>>> >>>> >>>> https://github.com/sprabhakar2006/openSCAD/blob/main/multiple_sec_extrude_example2.scad >>>> >>>> >>>> https://github.com/sprabhakar2006/openSCAD/blob/main/multiple_section_extrude.ipynb >>>> [image: Screenshot 2022-10-16 at 9.35.54 PM.png] >>>> >>>> On Sun, 16 Oct 2022 at 06:45, Sanjeev Prabhakar < >>>> sprabhakar2006@gmail.com> wrote: >>>> >>>>> OK, if that is the case then it is a bit simpler >>>>> >>>>> On Sun, 16 Oct 2022 at 06:12, Adrian Mariano <avm4@cornell.edu> wrote: >>>>> >>>>>> neri-engineering is just talking about a basic sweep algorithm >>>>>> applied to a circular cross section, which length-dependent scale. It's >>>>>> what the original poster was donig with the neck1200 model. He was >>>>>> suggesting creating the path with beziers or some other method to produce a >>>>>> smooth path. Since he assumes a circular cross section, all the >>>>>> complications having to do with twisting disappear. >>>>>> >>>>>> On Sat, Oct 15, 2022 at 8:36 PM Sanjeev Prabhakar < >>>>>> sprabhakar2006@gmail.com> wrote: >>>>>> >>>>>>> Thanks for the explanation >>>>>>> Can you share a hand sketch of what will be the final expected >>>>>>> figure. >>>>>>> Maybe someone here has already written something similar. >>>>>>> >>>>>>> On Sun, 16 Oct, 2022, 12:55 am neri-engineering via Discuss, < >>>>>>> discuss@lists.openscad.org> wrote: >>>>>>> >>>>>>>> >>>>>>>> >>>>>>>> >>>>>>>> ---------- Forwarded message ---------- >>>>>>>> From: neri-engineering <neri-engineering@protonmail.com> >>>>>>>> To: OpenSCAD general discussion Mailing-list < >>>>>>>> discuss@lists.openscad.org> >>>>>>>> Cc: >>>>>>>> Bcc: >>>>>>>> Date: Sat, 15 Oct 2022 19:24:32 +0000 >>>>>>>> Subject: [OpenSCAD] Re: multiple section extrude >>>>>>>> Long ago I wrote something similar, in order to generate a Klein >>>>>>>> bottle, and for that reason I can offer a suggestion. >>>>>>>> >>>>>>>> The "path of center" could be a smooth curve, e.g. Bezier or cubic >>>>>>>> curve, where a discrete set of points defines the actual path (center curve >>>>>>>> passes through points) or the points serve as "control points". I forget >>>>>>>> the technical name for these sorts of curve, I only remember that there are >>>>>>>> several variations, and I think they involved some sort of cubic curves. >>>>>>>> >>>>>>>> The radius as it varies as you travel through the center curve >>>>>>>> could also be a smooth curve, following a similar strategy. Control points >>>>>>>> or actual radii at sample points could determine the radius along every >>>>>>>> point of the center curve. >>>>>>>> >>>>>>>> So in essence what I'm proposing are two "smooth" functions in >>>>>>>> terms of 't' which is in the interval [0,1]. >>>>>>>> >>>>>>>> First function C:[0,1] -> R3 maps every point in the interval [0,1] >>>>>>>> to a point in 3-space and the smooth curve defined such is the center of >>>>>>>> your "tube". >>>>>>>> >>>>>>>> Second function D:[0,1] -> R maps every point in the interval [0,1] >>>>>>>> to a real value, which is defined to be the radius of tube at that location. >>>>>>>> >>>>>>>> Of course you'd have to do things like figuring out how to trace >>>>>>>> the circle at every point, based on some simple vector calculations, to >>>>>>>> determine perpendiculars etc. Simple vector calculations, dot products, >>>>>>>> cross products, and the like. >>>>>>>> >>>>>>>> You'd get a very smooth tube which would be highly desirable. >>>>>>>> >>>>>>>> I'm only sharing this because I had great success with the idea. >>>>>>>> >>>>>>>> https://odysee.com/@Neri_Engineering/cyclist-on-klein-bottle >>>>>>>> >>>>>>>> Sent with Proton Mail <https://proton.me/> secure email. >>>>>>>> >>>>>>>> ------- Original Message ------- >>>>>>>> On Saturday, October 15th, 2022 at 12:04 PM, Sanjeev Prabhakar < >>>>>>>> sprabhakar2006@gmail.com> wrote: >>>>>>>> >>>>>>>> I wrote a function "multiple section extrude" >>>>>>>> >>>>>>>> Idea is supply path points, sections for each path points and >>>>>>>> radiuses for each path point. This will create a polyhedron with >>>>>>>> multi_sections like the example picture shown below. >>>>>>>> >>>>>>>> I am sure the similar function can be written in original openscad >>>>>>>> >>>>>>>> example: >>>>>>>> >>>>>>>> https://github.com/sprabhakar2006/openSCAD/blob/main/multiple_section_extrude.ipynb >>>>>>>> >>>>>>>> >>>>>>>> https://github.com/sprabhakar2006/openSCAD/blob/main/multiple_section_extrude.scad >>>>>>>> [image: Screenshot 2022-10-15 at 10.14.37 PM.png] >>>>>>>> >>>>>>>> >>>>>>>> >>>>>>>> >>>>>>>> >>>>>>>> ---------- Forwarded message ---------- >>>>>>>> From: neri-engineering via Discuss <discuss@lists.openscad.org> >>>>>>>> To: OpenSCAD general discussion Mailing-list < >>>>>>>> discuss@lists.openscad.org> >>>>>>>> Cc: neri-engineering <neri-engineering@protonmail.com> >>>>>>>> Bcc: >>>>>>>> Date: Sat, 15 Oct 2022 19:24:32 +0000 >>>>>>>> Subject: [OpenSCAD] Re: multiple section extrude >>>>>>>> _______________________________________________ >>>>>>>> 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 >

OK I was sharing various applications for multiple section extrude, so this is one of them. Method you are proposing is a good one On Mon, 17 Oct, 2022, 9:56 am Adrian Mariano, <avm4@cornell.edu> wrote: > I think I understand, but I had a different approach for how to make the > shape. You don't need to build this from the curved section 2 that you > show, because you can create it as a series of slices of a rectangle: a > scaled path_sweep. I tried to roughly duplicate the shape using this idea > in BOSL2. (Note: this is unreleased code so it won't run yet on the posted > BOSL2, which doesn't support scaling.) > > opath = [[0,0,0], [100,0,0], [200, 0, 33]]; > path=subdivide_path(opath, 100,closed=false); > scales = [each repeat([1,1],50), > for(i=[1:-1/50:1/50]) [sqrt(i),1] > ]; > path_sweep(rect([150,10]),path,scale=scales); > > The point is that the shape is just a rectangle scaled in its width with a > continuously decreasing width that approaches zero along a curved path for > the end. > > [image: image.png] > > > > On Sun, Oct 16, 2022 at 9:16 PM Sanjeev Prabhakar < > sprabhakar2006@gmail.com> wrote: > >> for better understanding, see the picture >> [image: Screenshot 2022-10-17 at 6.39.01 AM.png] >> >> On Sun, 16 Oct 2022 at 22:41, Sanjeev Prabhakar <sprabhakar2006@gmail.com> >> wrote: >> >>> No it's not scale >>> >>> There are 2 3d sections >>> >>> Both the sections have same number of points >>> >>> If you see the sections from top view both are rectangular, 1 larger >>> and 1 smaller >>> >>> If you view the sections from " -y" the larger section will be almost a >>> straight line and smaller section will be an arc (arc pointing on bottom >>> side). >>> >>> I hope this could explain to some extent. >>> >>> You can read the code for better understanding. >>> >>> >>> >>> >>> >>> On Sun, 16 Oct, 2022, 10:02 pm Adrian Mariano, <avm4@cornell.edu> wrote: >>> >>>> The section changes scale in one axis, but otherwise remains the same, >>>> no? >>>> >>>> On Sun, Oct 16, 2022 at 12:16 PM Sanjeev Prabhakar < >>>> sprabhakar2006@gmail.com> wrote: >>>> >>>>> another example of multiple section extrude function. >>>>> It might seem trivial, but it is not. >>>>> >>>>> >>>>> https://github.com/sprabhakar2006/openSCAD/blob/main/multiple_sec_extrude_example2.scad >>>>> >>>>> >>>>> https://github.com/sprabhakar2006/openSCAD/blob/main/multiple_section_extrude.ipynb >>>>> [image: Screenshot 2022-10-16 at 9.35.54 PM.png] >>>>> >>>>> On Sun, 16 Oct 2022 at 06:45, Sanjeev Prabhakar < >>>>> sprabhakar2006@gmail.com> wrote: >>>>> >>>>>> OK, if that is the case then it is a bit simpler >>>>>> >>>>>> On Sun, 16 Oct 2022 at 06:12, Adrian Mariano <avm4@cornell.edu> >>>>>> wrote: >>>>>> >>>>>>> neri-engineering is just talking about a basic sweep algorithm >>>>>>> applied to a circular cross section, which length-dependent scale. It's >>>>>>> what the original poster was donig with the neck1200 model. He was >>>>>>> suggesting creating the path with beziers or some other method to produce a >>>>>>> smooth path. Since he assumes a circular cross section, all the >>>>>>> complications having to do with twisting disappear. >>>>>>> >>>>>>> On Sat, Oct 15, 2022 at 8:36 PM Sanjeev Prabhakar < >>>>>>> sprabhakar2006@gmail.com> wrote: >>>>>>> >>>>>>>> Thanks for the explanation >>>>>>>> Can you share a hand sketch of what will be the final expected >>>>>>>> figure. >>>>>>>> Maybe someone here has already written something similar. >>>>>>>> >>>>>>>> On Sun, 16 Oct, 2022, 12:55 am neri-engineering via Discuss, < >>>>>>>> discuss@lists.openscad.org> wrote: >>>>>>>> >>>>>>>>> >>>>>>>>> >>>>>>>>> >>>>>>>>> ---------- Forwarded message ---------- >>>>>>>>> From: neri-engineering <neri-engineering@protonmail.com> >>>>>>>>> To: OpenSCAD general discussion Mailing-list < >>>>>>>>> discuss@lists.openscad.org> >>>>>>>>> Cc: >>>>>>>>> Bcc: >>>>>>>>> Date: Sat, 15 Oct 2022 19:24:32 +0000 >>>>>>>>> Subject: [OpenSCAD] Re: multiple section extrude >>>>>>>>> Long ago I wrote something similar, in order to generate a Klein >>>>>>>>> bottle, and for that reason I can offer a suggestion. >>>>>>>>> >>>>>>>>> The "path of center" could be a smooth curve, e.g. Bezier or cubic >>>>>>>>> curve, where a discrete set of points defines the actual path (center curve >>>>>>>>> passes through points) or the points serve as "control points". I forget >>>>>>>>> the technical name for these sorts of curve, I only remember that there are >>>>>>>>> several variations, and I think they involved some sort of cubic curves. >>>>>>>>> >>>>>>>>> The radius as it varies as you travel through the center curve >>>>>>>>> could also be a smooth curve, following a similar strategy. Control points >>>>>>>>> or actual radii at sample points could determine the radius along every >>>>>>>>> point of the center curve. >>>>>>>>> >>>>>>>>> So in essence what I'm proposing are two "smooth" functions in >>>>>>>>> terms of 't' which is in the interval [0,1]. >>>>>>>>> >>>>>>>>> First function C:[0,1] -> R3 maps every point in the interval >>>>>>>>> [0,1] to a point in 3-space and the smooth curve defined such is the center >>>>>>>>> of your "tube". >>>>>>>>> >>>>>>>>> Second function D:[0,1] -> R maps every point in the interval >>>>>>>>> [0,1] to a real value, which is defined to be the radius of tube at that >>>>>>>>> location. >>>>>>>>> >>>>>>>>> Of course you'd have to do things like figuring out how to trace >>>>>>>>> the circle at every point, based on some simple vector calculations, to >>>>>>>>> determine perpendiculars etc. Simple vector calculations, dot products, >>>>>>>>> cross products, and the like. >>>>>>>>> >>>>>>>>> You'd get a very smooth tube which would be highly desirable. >>>>>>>>> >>>>>>>>> I'm only sharing this because I had great success with the idea. >>>>>>>>> >>>>>>>>> https://odysee.com/@Neri_Engineering/cyclist-on-klein-bottle >>>>>>>>> >>>>>>>>> Sent with Proton Mail <https://proton.me/> secure email. >>>>>>>>> >>>>>>>>> ------- Original Message ------- >>>>>>>>> On Saturday, October 15th, 2022 at 12:04 PM, Sanjeev Prabhakar < >>>>>>>>> sprabhakar2006@gmail.com> wrote: >>>>>>>>> >>>>>>>>> I wrote a function "multiple section extrude" >>>>>>>>> >>>>>>>>> Idea is supply path points, sections for each path points and >>>>>>>>> radiuses for each path point. This will create a polyhedron with >>>>>>>>> multi_sections like the example picture shown below. >>>>>>>>> >>>>>>>>> I am sure the similar function can be written in original openscad >>>>>>>>> >>>>>>>>> example: >>>>>>>>> >>>>>>>>> https://github.com/sprabhakar2006/openSCAD/blob/main/multiple_section_extrude.ipynb >>>>>>>>> >>>>>>>>> >>>>>>>>> https://github.com/sprabhakar2006/openSCAD/blob/main/multiple_section_extrude.scad >>>>>>>>> [image: Screenshot 2022-10-15 at 10.14.37 PM.png] >>>>>>>>> >>>>>>>>> >>>>>>>>> >>>>>>>>> >>>>>>>>> >>>>>>>>> ---------- Forwarded message ---------- >>>>>>>>> From: neri-engineering via Discuss <discuss@lists.openscad.org> >>>>>>>>> To: OpenSCAD general discussion Mailing-list < >>>>>>>>> discuss@lists.openscad.org> >>>>>>>>> Cc: neri-engineering <neri-engineering@protonmail.com> >>>>>>>>> Bcc: >>>>>>>>> Date: Sat, 15 Oct 2022 19:24:32 +0000 >>>>>>>>> Subject: [OpenSCAD] Re: multiple section extrude >>>>>>>>> _______________________________________________ >>>>>>>>> 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 >

another example of multiple section extrude https://github.com/sprabhakar2006/openSCAD/blob/main/multiple_sec_extrude_ex3.scad [image: Screenshot 2022-10-22 at 1.18.25 PM.png] On Mon, 17 Oct 2022 at 10:41, Sanjeev Prabhakar <sprabhakar2006@gmail.com> wrote: > OK > I was sharing various applications for multiple section extrude, so this > is one of them. > Method you are proposing is a good one > > > On Mon, 17 Oct, 2022, 9:56 am Adrian Mariano, <avm4@cornell.edu> wrote: > >> I think I understand, but I had a different approach for how to make the >> shape. You don't need to build this from the curved section 2 that you >> show, because you can create it as a series of slices of a rectangle: a >> scaled path_sweep. I tried to roughly duplicate the shape using this idea >> in BOSL2. (Note: this is unreleased code so it won't run yet on the posted >> BOSL2, which doesn't support scaling.) >> >> opath = [[0,0,0], [100,0,0], [200, 0, 33]]; >> path=subdivide_path(opath, 100,closed=false); >> scales = [each repeat([1,1],50), >> for(i=[1:-1/50:1/50]) [sqrt(i),1] >> ]; >> path_sweep(rect([150,10]),path,scale=scales); >> >> The point is that the shape is just a rectangle scaled in its width with >> a continuously decreasing width that approaches zero along a curved path >> for the end. >> >> [image: image.png] >> >> >> >> On Sun, Oct 16, 2022 at 9:16 PM Sanjeev Prabhakar < >> sprabhakar2006@gmail.com> wrote: >> >>> for better understanding, see the picture >>> [image: Screenshot 2022-10-17 at 6.39.01 AM.png] >>> >>> On Sun, 16 Oct 2022 at 22:41, Sanjeev Prabhakar < >>> sprabhakar2006@gmail.com> wrote: >>> >>>> No it's not scale >>>> >>>> There are 2 3d sections >>>> >>>> Both the sections have same number of points >>>> >>>> If you see the sections from top view both are rectangular, 1 larger >>>> and 1 smaller >>>> >>>> If you view the sections from " -y" the larger section will be almost >>>> a straight line and smaller section will be an arc (arc pointing on bottom >>>> side). >>>> >>>> I hope this could explain to some extent. >>>> >>>> You can read the code for better understanding. >>>> >>>> >>>> >>>> >>>> >>>> On Sun, 16 Oct, 2022, 10:02 pm Adrian Mariano, <avm4@cornell.edu> >>>> wrote: >>>> >>>>> The section changes scale in one axis, but otherwise remains the same, >>>>> no? >>>>> >>>>> On Sun, Oct 16, 2022 at 12:16 PM Sanjeev Prabhakar < >>>>> sprabhakar2006@gmail.com> wrote: >>>>> >>>>>> another example of multiple section extrude function. >>>>>> It might seem trivial, but it is not. >>>>>> >>>>>> >>>>>> https://github.com/sprabhakar2006/openSCAD/blob/main/multiple_sec_extrude_example2.scad >>>>>> >>>>>> >>>>>> https://github.com/sprabhakar2006/openSCAD/blob/main/multiple_section_extrude.ipynb >>>>>> [image: Screenshot 2022-10-16 at 9.35.54 PM.png] >>>>>> >>>>>> On Sun, 16 Oct 2022 at 06:45, Sanjeev Prabhakar < >>>>>> sprabhakar2006@gmail.com> wrote: >>>>>> >>>>>>> OK, if that is the case then it is a bit simpler >>>>>>> >>>>>>> On Sun, 16 Oct 2022 at 06:12, Adrian Mariano <avm4@cornell.edu> >>>>>>> wrote: >>>>>>> >>>>>>>> neri-engineering is just talking about a basic sweep algorithm >>>>>>>> applied to a circular cross section, which length-dependent scale. It's >>>>>>>> what the original poster was donig with the neck1200 model. He was >>>>>>>> suggesting creating the path with beziers or some other method to produce a >>>>>>>> smooth path. Since he assumes a circular cross section, all the >>>>>>>> complications having to do with twisting disappear. >>>>>>>> >>>>>>>> On Sat, Oct 15, 2022 at 8:36 PM Sanjeev Prabhakar < >>>>>>>> sprabhakar2006@gmail.com> wrote: >>>>>>>> >>>>>>>>> Thanks for the explanation >>>>>>>>> Can you share a hand sketch of what will be the final expected >>>>>>>>> figure. >>>>>>>>> Maybe someone here has already written something similar. >>>>>>>>> >>>>>>>>> On Sun, 16 Oct, 2022, 12:55 am neri-engineering via Discuss, < >>>>>>>>> discuss@lists.openscad.org> wrote: >>>>>>>>> >>>>>>>>>> >>>>>>>>>> >>>>>>>>>> >>>>>>>>>> ---------- Forwarded message ---------- >>>>>>>>>> From: neri-engineering <neri-engineering@protonmail.com> >>>>>>>>>> To: OpenSCAD general discussion Mailing-list < >>>>>>>>>> discuss@lists.openscad.org> >>>>>>>>>> Cc: >>>>>>>>>> Bcc: >>>>>>>>>> Date: Sat, 15 Oct 2022 19:24:32 +0000 >>>>>>>>>> Subject: [OpenSCAD] Re: multiple section extrude >>>>>>>>>> Long ago I wrote something similar, in order to generate a Klein >>>>>>>>>> bottle, and for that reason I can offer a suggestion. >>>>>>>>>> >>>>>>>>>> The "path of center" could be a smooth curve, e.g. Bezier or >>>>>>>>>> cubic curve, where a discrete set of points defines the actual path (center >>>>>>>>>> curve passes through points) or the points serve as "control points". I >>>>>>>>>> forget the technical name for these sorts of curve, I only remember that >>>>>>>>>> there are several variations, and I think they involved some sort of cubic >>>>>>>>>> curves. >>>>>>>>>> >>>>>>>>>> The radius as it varies as you travel through the center curve >>>>>>>>>> could also be a smooth curve, following a similar strategy. Control points >>>>>>>>>> or actual radii at sample points could determine the radius along every >>>>>>>>>> point of the center curve. >>>>>>>>>> >>>>>>>>>> So in essence what I'm proposing are two "smooth" functions in >>>>>>>>>> terms of 't' which is in the interval [0,1]. >>>>>>>>>> >>>>>>>>>> First function C:[0,1] -> R3 maps every point in the interval >>>>>>>>>> [0,1] to a point in 3-space and the smooth curve defined such is the center >>>>>>>>>> of your "tube". >>>>>>>>>> >>>>>>>>>> Second function D:[0,1] -> R maps every point in the interval >>>>>>>>>> [0,1] to a real value, which is defined to be the radius of tube at that >>>>>>>>>> location. >>>>>>>>>> >>>>>>>>>> Of course you'd have to do things like figuring out how to trace >>>>>>>>>> the circle at every point, based on some simple vector calculations, to >>>>>>>>>> determine perpendiculars etc. Simple vector calculations, dot products, >>>>>>>>>> cross products, and the like. >>>>>>>>>> >>>>>>>>>> You'd get a very smooth tube which would be highly desirable. >>>>>>>>>> >>>>>>>>>> I'm only sharing this because I had great success with the idea. >>>>>>>>>> >>>>>>>>>> https://odysee.com/@Neri_Engineering/cyclist-on-klein-bottle >>>>>>>>>> >>>>>>>>>> Sent with Proton Mail <https://proton.me/> secure email. >>>>>>>>>> >>>>>>>>>> ------- Original Message ------- >>>>>>>>>> On Saturday, October 15th, 2022 at 12:04 PM, Sanjeev Prabhakar < >>>>>>>>>> sprabhakar2006@gmail.com> wrote: >>>>>>>>>> >>>>>>>>>> I wrote a function "multiple section extrude" >>>>>>>>>> >>>>>>>>>> Idea is supply path points, sections for each path points and >>>>>>>>>> radiuses for each path point. This will create a polyhedron with >>>>>>>>>> multi_sections like the example picture shown below. >>>>>>>>>> >>>>>>>>>> I am sure the similar function can be written in original openscad >>>>>>>>>> >>>>>>>>>> example: >>>>>>>>>> >>>>>>>>>> https://github.com/sprabhakar2006/openSCAD/blob/main/multiple_section_extrude.ipynb >>>>>>>>>> >>>>>>>>>> >>>>>>>>>> https://github.com/sprabhakar2006/openSCAD/blob/main/multiple_section_extrude.scad >>>>>>>>>> [image: Screenshot 2022-10-15 at 10.14.37 PM.png] >>>>>>>>>> >>>>>>>>>> >>>>>>>>>> >>>>>>>>>> >>>>>>>>>> >>>>>>>>>> ---------- Forwarded message ---------- >>>>>>>>>> From: neri-engineering via Discuss <discuss@lists.openscad.org> >>>>>>>>>> To: OpenSCAD general discussion Mailing-list < >>>>>>>>>> discuss@lists.openscad.org> >>>>>>>>>> Cc: neri-engineering <neri-engineering@protonmail.com> >>>>>>>>>> Bcc: >>>>>>>>>> Date: Sat, 15 Oct 2022 19:24:32 +0000 >>>>>>>>>> Subject: [OpenSCAD] Re: multiple section extrude >>>>>>>>>> _______________________________________________ >>>>>>>>>> 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 >> >