Made a sphere

[Bill Kahn]

Actually, I can share my OnShape document. All you need is a free account and you can tweak the parameters to suit. No prior knowledge required, just put you numbers in and the drawing will update itself. If anyone is interested, I'll do a little video explanation. Stefan's was great, but he missed some of the basics like setting the cut radius and lining up. I'll get on it

Thank you so much. Yes, I have an OnShape account and have played with it a bit. Have enjoyed it. I will learn a lot in looking how you set this up. Thanks. -Bill

 

[Janderso]

Maybe I missed it, what angle are you putting the spin indexer device?
Is it even that critical?

 

[Bill Kahn]

Maybe I missed it, what angle are you putting the spin indexer device?
Is it even that critical?

What I do: Let d=diameter of the tab. D be the diameter of the cylinder. The angle I use is atan(d/D). At this angle the cutter is just starting to cut the top of the tab at the same height as it is (finally) also cutting the very middle of the cylinder. If the angle is less than this (a flatter cylinder) then the cutter starts to cut into the tab before it has quite gotten low enough to round the middle of the cylinder. Connected to this is how long the cylinder is. What I do is L=D/cos(angle). So, if you set the angle to be greater than angle you find your cylinder is not long enough and to save the piece you raise the boring bar, reduce D, and come down again. (Yeah, ask me how I know.) Effectively you are back solving for D given the (too large) angle you did set.

Now, all this being said, I am not a geometrician. So this is just me noodling and doodling, and having now made 6 of these (3 not working out for various reasons). If anyone on the list can do the math more carefully, I would love to see the correct solution. (https://meta.wikimedia.org/wiki/Cunningham's_Law Cunningham's Law)

-Bill

 

[Janderso]

Gosh Bill,
I’ll need to noodle and doodle a bit to work out your formula.
You make it look easy.

 

[Bill Kahn]

My 7th sphere now made. Aluminum this time, so, no antimicrobe properties. BUT, this one is 3.4” in diameter. Which is much much bigger than the 1” and 1.2” ones I had been making. Of course, a picture of a sphere of any size always looks the same, but here is a snapshot anyhow. ThE size required some adjustments to my method—nothing big, but it is interesting how some machining operations do not scale. This sphere came out more spherical than any of the others. 3.400” +/- .002”. So, far from ball bearing quality, but rolls on the surface plate just like a ball bearing. No way to tell it isn’t a sphere by feeling it, watching it roll, or looking at the reflections. You really do need (long jaw) calipers.

You can see I worked on the surface finish. Aluminum polishes up nicely. For the final polish Mother’s Billet polish worked well. The snap shot catches the overhead recessed lights, which are close to point sources. And you can see diffraction spikes coming out in a few directions. This indicates that on one of the (many) sanding steps, I did not take enough time to get the previous sanding gouges out. Actually, I think it was the very last step, going from 2500 to 3000 grit. So, it would not really be that much work to take it back to 3000 again and repolish. But, this is a sort of geeky thing. If you are not looking for diffraction spikes off of point sources, it is really quite shiny. And, maybe a small point of physics instruction someday for some teen-aged aspiring scientist.

So, (in addition to many other things) I have now machined all five Platonic solids. And spheres. Three years in now and still loving this hobby.

-Bill

 

[Lo-Fi]

Fantastic! That finish is incredible!

Sorry I've not got round to posting up drawings and suchlike yet, I've been way too wrapped up in a little project myself. I'll get something sorted over the weekend.