How to model a golf ball?? I have no clue

Charless

Charless

0
Messages
449
From the net:
"If you are counting, most regulation golf balls will fall into the 300 to 450 dimple range. If you are measuring, your typical dint is about 7/1000th of an inch deep. Engineers say every 1/1000th is worth about five yards on the course."
 
* Some quick poking in the Wikipedia implies that the geometry of dimples is based on an icosahedron. So, start with that and add a Stam-Loop subdivider as you require triangulation.
* Create dimples on the points by inner extrusion and normal move. Add another subdivider.
* Actually, my icosahedral theory seems a bit :sick: fishy (as are most of my theories). There is an image of a Slazenger ball in the WP which certainly is not an icoshedron.
:rolleyes: I know nothing about the aerodynamics, but the fine-tuning of the dimply method should be simple.

GolfBall.png
 
* Almost:
:oops: Hexaganons don´t tesselate on a sphere. The have cleverly hidden some pentagonal dimples at the apices of the icosahedron.
Callaway.png
 
* Some further research:
* U.S. Pat. No. 4,560,168 describes an icosahedral dimple pattern. The dimples are positioned within the spherical icosahedral triangles so that the dimples do not intersect the six great circles which pass through the midpoints of the sides of the triangles.
* U.S. Pat. No. 4,142,227 describes a dodecahedral dimple pattern which includes 10 great circles which do not intersect dimples. However, the surface of the ball includes from 12 to 30 rectangular bald patches or dimple-free areas.

* Icosahedral base: As per the US patent 4,560,168 my method in #2 is :poop: Quatsch. Dimples go to the faces and not to the points. All faces, of course, are the subdivided triangles of the basic icosahedron and all dimples are of the same size. By definitions (going back to Platonic solids), there are 12 pentagonal clusters of dimples.
* As mentioned by ZooHead, manufacturers use different geometries.
 
Helmut said:
* Almost:
:oops: Hexaganons don´t tesselate on a sphere. The have cleverly hidden some pentagonal dimples at the apices of the icosahedron.
Click to expand...
Good eye, I noticed that on the image I posted, after I posted it.
 
* Not so much a :rolleyes: good eye: Long ago, I was fascinated by Buckminster Fullers geodesic domes, and - much later - I modelled C60 Buckminsterfullerenes / Buckyballs. Molecular geometry is great - albeit infinitesimal - stuff.
* As most C3D regulars may know, I am a geometry nerd, be this subatomic particles, molecules, Hagia Sophia to Zaha Hadid or galactic metaclusters.
 
@Helmut I helped build a 40 foot dome in my youth, and then I would
sit atop a ladder and play my flute as there was an incredible echo.

@Charless You picked a tough one. I'm getting close with a Particle Mesh and an Icosahedron with Boolean dimples.
 
There's a small problem that can be fixed.

Linear Subdivision divides pentagons in such a way to
disrupt the golf ball pattern slightly but it's quite noticeable.

You can see how the pattern is changed.

Thick as a brick: I should have been using an Icosohedron, not an Icosohedron truncated. :sleep:

golf-ball-dimple.gif
 
Last edited:
Icosahedron+stamloop+spherify-modifier; the dimple-polygons are beveled points till it reaches an almost evenly distribution of 470 hexagons +12 pentagons - inner-extrusion with shift-key followed by a Laplacian-smoothing:
Golfen.png
 
Last edited:
Helmut said:
* Some quick poking in the Wikipedia implies that the geometry of dimples is based on an icosahedron. So, start with that and add a Stam-Loop subdivider as you require triangulation.
* Create dimples on the points by inner extrusion and normal move. Add another subdivider.
Click to expand...
Helmut got it right earlier I'm seeing now. (y)
I was too thick to see it before.

Thanks to both Helmut and EllenM for their expertise.
I was going in the wrong direction again. :sleep:
 
My method creates 642 n-gons/dimples, which is more than the originally stated 300 to 450.
Not sure if it even maters as it still looks correct. Maybe if I was working for a manufacturer.
But in that case I probably wouldn't be modeling a golf ball, I'd be photographing one.
 
I´m still experimenting and came across this topology:
RoundDimples.jpeg

Pretty to look at - but unfortunately impossible to fabricate as it can't be split in half without a noticeable seam I guess.
 
* Returning to @EllenM ´s original suggestion (which she has deleted):
:oops: It is possible to generate a dimply golf ball based on a cube. I have used a tetrakis hexahedron. No idea about the aerodynamics of this option, but manufacturers will have analysed all that long ago.
* There are great circles in this topology, so injection moulding that in two parts should be possible.

dimps 1.png
 
* Mind you, my only exposure to golf balls was a brief scene where James Bond (†) and Mr Goldfinger (†) displayed their skills. :sick: Somewhere in the last century, even in the last millennium.
 
Here's how I did it. You can use a Spherify Modifier at the end.
After this use Inner Extrude with Preserve Group unchecked.
Sometimes deleting the edges also deletes some polygons, don't know why.

golf_ball.gif
 
You must log in or register to reply here.
Back
Top