modeling a complex object
I was asked to model the handle shown in this brief video:
I did a 3D scan, which produced a recognizable version of the handle,
but not anything I would give to a customer. Attempts at improving
things using sculpting software (SculptGL) helped a bit, but the scan
seems to have a lot of internal errors (voids inside the object) which
were revealed when I sliced the scan for printing.
It may well be that modeling this in OS is impossible, but I wondered if
anyone had an approach that they would suggest. The handle need not be
an exact replica, but at least something that is not embarrassing. Of
course, the so-created handle could well interfere with other parts of
the assembly (that is, the exact shape may be due to parts around the
actuation path of the handle).
Note that the counterbore at the non-pointy end, and the six slanted
nubs and the through hole are easy enough to model. It is the graceful
shape that eludes me.
Thanks!
Jon
--
This email has been checked for viruses by AVG antivirus software.
www.avg.com
On Saturday, May 30, 2026 at 12:32:53 PM EDT, Jon Bondy via Discuss discuss@lists.openscad.org wrote:
My inclination if using OpenSCAD would be to measure the extremities and position suitably sized spheres and hull() them together, repeating this for each feature/angle/portion of perimeter.
Iterate and adjust on that until you arrive at something workable?
William
--
Sphinx of black quartz, judge my vow.
https://designinto3d.com/
Anno domini 16:36:31 Sat, 30 May 2026 +0000 (UTC)
William F. Adams via Discuss scripsit:
On Saturday, May 30, 2026 at 12:32:53 PM EDT, Jon Bondy via Discuss discuss@lists.openscad.org wrote:
My inclination if using OpenSCAD would be to measure the extremities and position suitably sized spheres and hull() them together, repeating this for each feature/angle/portion of perimeter.
Iterate and adjust on that until you arrive at something workable?
As you already have the 3D scan you import it, make it semitransparent and use it to guide you in the process.
Nik
William
--
Sphinx of black quartz, judge my vow.
https://designinto3d.com/
OpenSCAD mailing list
To unsubscribe send an email to discuss-leave@lists.openscad.org
--
Please do not email me anything that you are not comfortable also sharing with the NSA, CIA ...
HI Jon,
If you've got it scanned in, good enough that you thought it would slice
OK, then i guess you have the shape. I would use Meshlab, or something,
to remove the voids. You do not need to learn Meshlab, just ask an AI,
to solve that particular problem.
If I wanted the fun of using Openscad, then I would take maybe half a
dozen photos, with either a ruler, or graphpaper in background, then use
a photo editor to get the object images to the same size.
A good start, is then to use a top view of the object, and get the 2d
points, either measure, or profile tracing/whatever., then the same for
a side view. Linear extrude both, rotate one 90degrees, and intersect.
There will be a lot of tidying up. If you measured the angles at which
you took the other photos, then add those extrudes too.
As you like writing software, then make or buy a Renishaw type probe,
and fix to your 3d printer. It has been out of patent now, for a number
of years, and ones 'good enough' are not too expensive, but interesting
to make for the sort of accuracy that is good enough. That will create a
useful point cloud, one you can work with and control.
Best wishes,
Ray
On 30/05/2026 17:32, Jon Bondy via Discuss wrote:
I was asked to model the handle shown in this brief video:
I did a 3D scan, which produced a recognizable version of the handle,
but not anything I would give to a customer. Attempts at improving
things using sculpting software (SculptGL) helped a bit, but the scan
seems to have a lot of internal errors (voids inside the object) which
were revealed when I sliced the scan for printing.
It may well be that modeling this in OS is impossible, but I wondered
if anyone had an approach that they would suggest. The handle need not
be an exact replica, but at least something that is not embarrassing.
Of course, the so-created handle could well interfere with other parts
of the assembly (that is, the exact shape may be due to parts around
the actuation path of the handle).
Note that the counterbore at the non-pointy end, and the six slanted
nubs and the through hole are easy enough to model. It is the
graceful shape that eludes me.
Thanks!
Jon
Not fun at all. I would look at stitching together Bezier patches.
https://github.com/BelfrySCAD/BOSL2/wiki/beziers.scad#section-bezier-surfaces
Hi Jon,
Can you post some still pictures of this handle?
I want to give it a try this weekend
Regards
On Sat, 30 May 2026 at 22:03, Jon Bondy via Discuss <
discuss@lists.openscad.org> wrote:
I was asked to model the handle shown in this brief video:
I did a 3D scan, which produced a recognizable version of the handle,
but not anything I would give to a customer. Attempts at improving
things using sculpting software (SculptGL) helped a bit, but the scan
seems to have a lot of internal errors (voids inside the object) which
were revealed when I sliced the scan for printing.
It may well be that modeling this in OS is impossible, but I wondered if
anyone had an approach that they would suggest. The handle need not be
an exact replica, but at least something that is not embarrassing. Of
course, the so-created handle could well interfere with other parts of
the assembly (that is, the exact shape may be due to parts around the
actuation path of the handle).
Note that the counterbore at the non-pointy end, and the six slanted
nubs and the through hole are easy enough to model. It is the graceful
shape that eludes me.
Thanks!
Jon
--
This email has been checked for viruses by AVG antivirus software.
www.avg.com
OpenSCAD mailing list
To unsubscribe send an email to discuss-leave@lists.openscad.org
The part is on its way back to its owner. All I can offer now is that you download the YouTube video and do some screen frame grabs.
Jon
On 6/2/2026 11:00 AM, Sanjeev Prabhakar wrote:
Hi Jon,
Can you post some still pictures of this handle?
I want to give it a try this weekend
Regards
On Sat, 30 May 2026 at 22:03, Jon Bondy via Discuss <discuss@lists.openscad.orgmailto:discuss@lists.openscad.org> wrote:
I was asked to model the handle shown in this brief video:
https://youtu.be/2Cl6_5Co-L4https://urldefense.proofpoint.com/v2/url?u=https-3A__youtu.be_2Cl6-5F5Co-2DL4&d=DwMFaQ&c=euGZstcaTDllvimEN8b7jXrwqOf-v5A_CdpgnVfiiMM&r=AsrE-c7ZR7B2Kyr3qgfvvppkCEBVsNmwEMndcrRSuOI&m=PCh_CDD_4ZaWo2eqN5GZU2oDJPZRmy4wmw_LBJfS4y31P0aivtBb1TA-HPEy-XTe&s=RuNeweI0Jbo4vsEmcA9KQPXahlcgRZsdaWI3e_hOoKA&e=
I did a 3D scan, which produced a recognizable version of the handle,
but not anything I would give to a customer. Attempts at improving
things using sculpting software (SculptGL) helped a bit, but the scan
seems to have a lot of internal errors (voids inside the object) which
were revealed when I sliced the scan for printing.
It may well be that modeling this in OS is impossible, but I wondered if
anyone had an approach that they would suggest. The handle need not be
an exact replica, but at least something that is not embarrassing. Of
course, the so-created handle could well interfere with other parts of
the assembly (that is, the exact shape may be due to parts around the
actuation path of the handle).
Note that the counterbore at the non-pointy end, and the six slanted
nubs and the through hole are easy enough to model. It is the graceful
shape that eludes me.
Thanks!
Jon
--
This email has been checked for viruses by AVG antivirus software.
www.avg.comhttps://urldefense.proofpoint.com/v2/url?u=http-3A__www.avg.com&d=DwMFaQ&c=euGZstcaTDllvimEN8b7jXrwqOf-v5A_CdpgnVfiiMM&r=AsrE-c7ZR7B2Kyr3qgfvvppkCEBVsNmwEMndcrRSuOI&m=PCh_CDD_4ZaWo2eqN5GZU2oDJPZRmy4wmw_LBJfS4y31P0aivtBb1TA-HPEy-XTe&s=aIG_B2Jt4cW6KKAzsay5ILK6SlSoSJ0jK6YfLJWn4OU&e=
OpenSCAD mailing list
To unsubscribe send an email to discuss-leave@lists.openscad.orgmailto:discuss-leave@lists.openscad.org
OK no problem
I will try
On Wed, 3 Jun, 2026, 3:59 am jon jonbondy.com, jon@jonbondy.com wrote:
The part is on its way back to its owner. All I can offer now is that you
download the YouTube video and do some screen frame grabs.
Jon
On 6/2/2026 11:00 AM, Sanjeev Prabhakar wrote:
Hi Jon,
Can you post some still pictures of this handle?
I want to give it a try this weekend
Regards
On Sat, 30 May 2026 at 22:03, Jon Bondy via Discuss <
discuss@lists.openscad.org> wrote:
I was asked to model the handle shown in this brief video:
https://youtu.be/2Cl6_5Co-L4
https://urldefense.proofpoint.com/v2/url?u=https-3A__youtu.be_2Cl6-5F5Co-2DL4&d=DwMFaQ&c=euGZstcaTDllvimEN8b7jXrwqOf-v5A_CdpgnVfiiMM&r=AsrE-c7ZR7B2Kyr3qgfvvppkCEBVsNmwEMndcrRSuOI&m=PCh_CDD_4ZaWo2eqN5GZU2oDJPZRmy4wmw_LBJfS4y31P0aivtBb1TA-HPEy-XTe&s=RuNeweI0Jbo4vsEmcA9KQPXahlcgRZsdaWI3e_hOoKA&e=
I did a 3D scan, which produced a recognizable version of the handle,
but not anything I would give to a customer. Attempts at improving
things using sculpting software (SculptGL) helped a bit, but the scan
seems to have a lot of internal errors (voids inside the object) which
were revealed when I sliced the scan for printing.
It may well be that modeling this in OS is impossible, but I wondered if
anyone had an approach that they would suggest. The handle need not be
an exact replica, but at least something that is not embarrassing. Of
course, the so-created handle could well interfere with other parts of
the assembly (that is, the exact shape may be due to parts around the
actuation path of the handle).
Note that the counterbore at the non-pointy end, and the six slanted
nubs and the through hole are easy enough to model. It is the graceful
shape that eludes me.
Thanks!
Jon
--
This email has been checked for viruses by AVG antivirus software.
www.avg.com
https://urldefense.proofpoint.com/v2/url?u=http-3A__www.avg.com&d=DwMFaQ&c=euGZstcaTDllvimEN8b7jXrwqOf-v5A_CdpgnVfiiMM&r=AsrE-c7ZR7B2Kyr3qgfvvppkCEBVsNmwEMndcrRSuOI&m=PCh_CDD_4ZaWo2eqN5GZU2oDJPZRmy4wmw_LBJfS4y31P0aivtBb1TA-HPEy-XTe&s=aIG_B2Jt4cW6KKAzsay5ILK6SlSoSJ0jK6YfLJWn4OU&e=
OpenSCAD mailing list
To unsubscribe send an email to discuss-leave@lists.openscad.org
I tried this today and it took a while
very difficult
posted the scad file: it is 1.6 mb (strange_shape.scad)
https://github.com/sprabhakar2006/openSCAD/blob/main/strange_shape.scad
associated_python code is attached
[image: Screenshot 2026-06-06 at 9.10.37 PM.png]
[image: Screenshot 2026-06-06 at 9.12.14 PM.png]
[image: Screenshot 2026-06-06 at 9.11.34 PM.png]
On Wed, 3 Jun 2026 at 06:03, Sanjeev Prabhakar sprabhakar2006@gmail.com
wrote:
OK no problem
I will try
On Wed, 3 Jun, 2026, 3:59 am jon jonbondy.com, jon@jonbondy.com wrote:
The part is on its way back to its owner. All I can offer now is that
you download the YouTube video and do some screen frame grabs.
Jon
On 6/2/2026 11:00 AM, Sanjeev Prabhakar wrote:
Hi Jon,
Can you post some still pictures of this handle?
I want to give it a try this weekend
Regards
On Sat, 30 May 2026 at 22:03, Jon Bondy via Discuss <
discuss@lists.openscad.org> wrote:
I was asked to model the handle shown in this brief video:
https://youtu.be/2Cl6_5Co-L4
https://urldefense.proofpoint.com/v2/url?u=https-3A__youtu.be_2Cl6-5F5Co-2DL4&d=DwMFaQ&c=euGZstcaTDllvimEN8b7jXrwqOf-v5A_CdpgnVfiiMM&r=AsrE-c7ZR7B2Kyr3qgfvvppkCEBVsNmwEMndcrRSuOI&m=PCh_CDD_4ZaWo2eqN5GZU2oDJPZRmy4wmw_LBJfS4y31P0aivtBb1TA-HPEy-XTe&s=RuNeweI0Jbo4vsEmcA9KQPXahlcgRZsdaWI3e_hOoKA&e=
I did a 3D scan, which produced a recognizable version of the handle,
but not anything I would give to a customer. Attempts at improving
things using sculpting software (SculptGL) helped a bit, but the scan
seems to have a lot of internal errors (voids inside the object) which
were revealed when I sliced the scan for printing.
It may well be that modeling this in OS is impossible, but I wondered if
anyone had an approach that they would suggest. The handle need not be
an exact replica, but at least something that is not embarrassing. Of
course, the so-created handle could well interfere with other parts of
the assembly (that is, the exact shape may be due to parts around the
actuation path of the handle).
Note that the counterbore at the non-pointy end, and the six slanted
nubs and the through hole are easy enough to model. It is the graceful
shape that eludes me.
Thanks!
Jon
--
This email has been checked for viruses by AVG antivirus software.
www.avg.com
https://urldefense.proofpoint.com/v2/url?u=http-3A__www.avg.com&d=DwMFaQ&c=euGZstcaTDllvimEN8b7jXrwqOf-v5A_CdpgnVfiiMM&r=AsrE-c7ZR7B2Kyr3qgfvvppkCEBVsNmwEMndcrRSuOI&m=PCh_CDD_4ZaWo2eqN5GZU2oDJPZRmy4wmw_LBJfS4y31P0aivtBb1TA-HPEy-XTe&s=aIG_B2Jt4cW6KKAzsay5ILK6SlSoSJ0jK6YfLJWn4OU&e=
OpenSCAD mailing list
To unsubscribe send an email to discuss-leave@lists.openscad.org
A very good likeness!
On 6/6/2026 12:00 PM, Sanjeev Prabhakar wrote:
I tried this today and it took a while
very difficult
posted the scad file: it is 1.6 mb (strange_shape.scad)
https://github.com/sprabhakar2006/openSCAD/blob/main/strange_shape.scadhttps://urldefense.proofpoint.com/v2/url?u=https-3A__github.com_sprabhakar2006_openSCAD_blob_main_strange-5Fshape.scad&d=DwMFaQ&c=euGZstcaTDllvimEN8b7jXrwqOf-v5A_CdpgnVfiiMM&r=AsrE-c7ZR7B2Kyr3qgfvvppkCEBVsNmwEMndcrRSuOI&m=9Vu-I-8lkbmP2hkkNkZhZEquxy-DGAf7E0uatGsET1tdBD4kP8Ltr6N0UmCTS-m4&s=FoY3gJudRGyPkmPBkF9EtixlL47PTgipiosfMTg4Id0&e=
associated_python code is attached
[Screenshot 2026-06-06 at 9.10.37 PM.png]
[Screenshot 2026-06-06 at 9.12.14 PM.png]
[Screenshot 2026-06-06 at 9.11.34 PM.png]
On Wed, 3 Jun 2026 at 06:03, Sanjeev Prabhakar <sprabhakar2006@gmail.commailto:sprabhakar2006@gmail.com> wrote:
OK no problem
I will try
On Wed, 3 Jun, 2026, 3:59 am jon jonbondy.comhttps://urldefense.proofpoint.com/v2/url?u=http-3A__jonbondy.com&d=DwMFaQ&c=euGZstcaTDllvimEN8b7jXrwqOf-v5A_CdpgnVfiiMM&r=AsrE-c7ZR7B2Kyr3qgfvvppkCEBVsNmwEMndcrRSuOI&m=9Vu-I-8lkbmP2hkkNkZhZEquxy-DGAf7E0uatGsET1tdBD4kP8Ltr6N0UmCTS-m4&s=zyMJSqOUqMC-3SbFgiS8uQv2MjufDTZmgcG47vNgfys&e=, <jon@jonbondy.commailto:jon@jonbondy.com> wrote:
The part is on its way back to its owner. All I can offer now is that you download the YouTube video and do some screen frame grabs.
Jon
On 6/2/2026 11:00 AM, Sanjeev Prabhakar wrote:
Hi Jon,
Can you post some still pictures of this handle?
I want to give it a try this weekend
Regards
On Sat, 30 May 2026 at 22:03, Jon Bondy via Discuss <discuss@lists.openscad.orgmailto:discuss@lists.openscad.org> wrote:
I was asked to model the handle shown in this brief video:
https://youtu.be/2Cl6_5Co-L4https://urldefense.proofpoint.com/v2/url?u=https-3A__youtu.be_2Cl6-5F5Co-2DL4&d=DwMFaQ&c=euGZstcaTDllvimEN8b7jXrwqOf-v5A_CdpgnVfiiMM&r=AsrE-c7ZR7B2Kyr3qgfvvppkCEBVsNmwEMndcrRSuOI&m=PCh_CDD_4ZaWo2eqN5GZU2oDJPZRmy4wmw_LBJfS4y31P0aivtBb1TA-HPEy-XTe&s=RuNeweI0Jbo4vsEmcA9KQPXahlcgRZsdaWI3e_hOoKA&e=
I did a 3D scan, which produced a recognizable version of the handle,
but not anything I would give to a customer. Attempts at improving
things using sculpting software (SculptGL) helped a bit, but the scan
seems to have a lot of internal errors (voids inside the object) which
were revealed when I sliced the scan for printing.
It may well be that modeling this in OS is impossible, but I wondered if
anyone had an approach that they would suggest. The handle need not be
an exact replica, but at least something that is not embarrassing. Of
course, the so-created handle could well interfere with other parts of
the assembly (that is, the exact shape may be due to parts around the
actuation path of the handle).
Note that the counterbore at the non-pointy end, and the six slanted
nubs and the through hole are easy enough to model. It is the graceful
shape that eludes me.
Thanks!
Jon
--
This email has been checked for viruses by AVG antivirus software.
www.avg.comhttps://urldefense.proofpoint.com/v2/url?u=http-3A__www.avg.com&d=DwMFaQ&c=euGZstcaTDllvimEN8b7jXrwqOf-v5A_CdpgnVfiiMM&r=AsrE-c7ZR7B2Kyr3qgfvvppkCEBVsNmwEMndcrRSuOI&m=PCh_CDD_4ZaWo2eqN5GZU2oDJPZRmy4wmw_LBJfS4y31P0aivtBb1TA-HPEy-XTe&s=aIG_B2Jt4cW6KKAzsay5ILK6SlSoSJ0jK6YfLJWn4OU&e=
OpenSCAD mailing list
To unsubscribe send an email to discuss-leave@lists.openscad.orgmailto:discuss-leave@lists.openscad.org
