Bevel gear: add spherical involute

[stefano2734]

Is your feature request related to a problem? Please describe.
A clear and concise description of what the problem is. Ex. I'm always frustrated when [...]

spherical involute is the best contour of a bevel gear with theoretical no or minimal slip at rolling test.
Straight bevel gear has no slip and spiral bevel gear minimal slip.

It was developed first in 1942 at Daimler Benz in Germany.

Describe the solution you'd like
A clear and concise description of what you want to happen.

See solution of 2006 by janssen for straight and helical bevel gears
https://www.thingiverse.com/thing:1588110

Higher difficulty is to implement all other parameters of a normal bevel gear.

Spiral bevel gear with spherical involute is also possible.

Describe alternatives you've considered
A clear and concise description of any alternative solutions or features you've considered.

Example Code
See downloads at
https://www.thingiverse.com/thing:1588110

// Code goes here.

Additional context
Add any other context or screenshots about the feature request here.

https://www.researchgate.net/publication/225659482_The_spherical_involute_bevel_gear_Its_geometry_kinematic_behavior_and_standardization

https://ijme.us/cd_11/PDF/Paper%20163%20ENG%20107.pdf

https://thermalprocessing.com/computerized-design-of-straight-bevel-gears-with-optimized-profiles-for-forging-molding-or-3d-printing/

[adrianVmariano]

Your text above is somewhat confusing with lots of boilerplate text still left in your post. It sounds like you are complaining that the octoid tooth form is not good enough and you want spherical involute? Is this really a significant issue? My understanding is that in most engineering applications, the octoid error is so small it has no measurable effect on performance, and machined gears are all octoid.

[stefano2734]

Your text above is somewhat confusing with lots of boilerplate text still left in your post. It sounds like you are complaining that the octoid tooth form is not good enough and you want spherical involute? Is this really a significant issue? My understanding is that in most engineering applications, the octoid error is so small it has no measurable effect on performance, and machined gears are all octoid.

Millions of spherical bevel gears are forged every year to differentials in millions of cars and trucks.
https://winningblw.com/produkte/

In 3D-printing with plastics or advanced in metals also spherical bevel gears are possible to improve old designs as spare parts.

Spherical involute is also used for spiral gears to improve the gears.
https://patents.google.com/patent/US20130216321A1/en

FVA Workbench includes this feature in version 10.0 at bevel gear.
https://doc.fva-service.de/knowledgebase/en/root-release-notes-100.html

So it is used in Industrie to increase lifetime by less transmission error with reduced friction and less contact stress.

And also the Chinese do it with a spherical involute Bézier curve.
https://engrxiv.org/preprint/view/6562

https://etagear.com/Articles/42

Patent: https://patents.google.com/patent/CN102371473A/en
https://patents.google.com/patent/CN108679196A/en
https://patents.google.com/patent/US9568085B2/en

https://patents.google.com/patent/EP2580493A2/de
https://patents.google.com/patent/EP2484474B1/de

[stefano2734]

A spherical involute Bézier curve is here a central element to to this feature quite simple.
A straight cylindrical involute curve goes from plane to sphere.

[adrianVmariano]

No, bezier curves are not capable of exactly representing a spherical involute. Nothing about gears is ever "quite simple" in my experience. Rather everything is quite complicated and then about ten times more complicated than you thought before. Your link salad above features many examples that show no indication they are anything but the standard octoid profile. The "Chinese" reference is about how spherical involutes introduce extra complexities that requires special treatment with bezier curves to strengthen the roots---not an example of "quite simple". I have every reason to believe that changing the existing code from octoid to spherical involute is many hours of effort, and you have not explained why it matters, given that the vast majority of gears used in practice are octoid anyway. I also don't think anybody is interested in pursuing this. Note that worm gears are broken and nobody knows how to fix them, whereas bevel gears are, as far as we know, working. If you want to convince us to keep this issue open for consideration in the distant future then you need to give a clear reason (not a random list of links) describing why it would be advantageous to change the existing code. How does the existing implementation limit applications? Why and how would anybody notice the change from octoid to spherical involute? What would it make possible that is currently not possible?

[stefano2734]

No, bezier curves are not capable of exactly representing a spherical involute. Nothing about gears is ever "quite simple" in my experience. Rather everything is quite complicated and then about ten times more complicated than you thought before. Your link salad above features many examples that show no indication they are anything but the standard octoid profile. The "Chinese" reference is about how spherical involutes introduce extra complexities that requires special treatment with bezier curves to strengthen the roots---not an example of "quite simple". I have every reason to believe that changing the existing code from octoid to spherical involute is many hours of effort, and you have not explained why it matters, given that the vast majority of gears used in practice are octoid anyway. I also don't think anybody is interested in pursuing this. Note that worm gears are broken and nobody knows how to fix them, whereas bevel gears are, as far as we know, working. If you want to convince us to keep this issue open for consideration in the distant future then you need to give a clear reason (not a random list of links) describing why it would be advantageous to change the existing code. How does the existing implementation limit applications? Why and how would anybody notice the change from octoid to spherical involute? What would it make possible that is currently not possible?

Chinese can do it: https://engrxiv.org/preprint/view/6562
It is like a landing on moon. So to heavy for Bosl2!?!
So do your other 10.000 issues you want before and 100.000 thanks for this and the actual state of bosl2 to all developers
and then perhaps in 100 years the difficult spherical involute

In short you can take a label: enhancement and or low priority or wontfix or others.

[adrianVmariano]

The moon landing was accomplished with a bit more than two people working for free only in their spare time. With limited time we have to prioritize and you are remarkably unable to give an actual reason why this is important, or what actual benefit it would provide. Repeatedly claiming that it's necessary, or criticizing us is not compelling. Feel free to contribute the desired feature. Note that your reference only describe straight bevel gears, it appears. What about zerol and spiral bevel gears?

[stefano2734]

The moon landing was accomplished with a bit more than two people working for free only in their spare time. With limited time we have to prioritize and you are remarkably unable to give an actual reason why this is important, or what actual benefit it would provide. Repeatedly claiming that it's necessary, or criticizing us is not compelling. Feel free to contribute the desired feature. Note that your reference only describe straight bevel gears, it appears. What about zerol and spiral bevel gears?

I would say it is also an enhancement like other issues with low priority.