View Full Version : Internal radius attachments?

10-24-2006, 10:37 PM
Okay everybody and the're brother has built or seen an external radius attachment for the lathe.

What I am needing is an internal radius attachment I.E. for turning spherical sockets(self-aligning bearing seats,ball joints etc.).

I have a few ideas,but was wondering if anyone has ever seen a commercial rig or built the're own homebrew?

10-24-2006, 11:16 PM
I've never seen one but if I had to do it on the fly, I'd grind a cutter and run it in just like you have to install/remove self-aligning bearings. Come in at an angle then swivel the piece when it reaches the center of the radius.

A ball socket usually has two sides affixed to capture a ball. Otherwise, how would you get the ball in the hole? :D Tooling is the biggest challenge. At least it would be for me.

10-25-2006, 12:01 AM
This is going to seem a little weird, but if you have an external radius attachment, you almost certainly have an internal radius attachment as well. It all depends on which side of the attachment's axis of rotation that the cutter is located. If the work is located between the cutting tool's axis of rotation and the cutting tool itself it will cut an external radius, a dome or ball centered on the axis of rotation. If the cutting tool or point is located between the work and the cutting tool's axis of rotation it will cut an internal radius, a bowl or socket with a radius equal to the distance between the cutting tool point and it's axis of rotation. I am sure there are more precise definitions, but this one makes the most sense to me.

Ian B
10-25-2006, 02:10 AM
The snag is going to be, how do you get the mechanism of an external radius attachment into the centre of the work - for a ball joint, the pivot point would normally need to be on the axis, smack in the middle of the spherical cavity that you want to machine.

This may work for larger radii, but gets awkward in smaller sizes.


10-25-2006, 07:57 AM
If you're talking very small, how 'bout grinding a drill to a radiused profile. Shouldn't be too hard to jig up even on a bench grinder.

John Stevenson
10-25-2006, 08:54 AM
'D' bits.

[ edit ] No the message isn't too short - get a life.

10-25-2006, 08:58 AM
I have successfully machined hemisheres on the ends of round bar stock using just a lathe and no special attachments. I wrote a program in Excel Spreadsheet to determine how much to increase the cross feed for each movement of the apron. I made very small incermental adjustments and finishing was done with a file to "smooth out" the very small steps used to make the cuts. I'm sure it can be modified to cut an interal hemisphere but I have not tried it. If interested I can provide the infromation.

10-25-2006, 10:52 AM
Almost all of the various types of radius attachments and other methods of generating radii will do concave radii and hemispheres as well. The problem is that with smaller sizes, there is not enough room for the tool or toolholder.

When you hit that point, a D bit is the easiest solution.

kap pullen
10-25-2006, 11:37 AM
Here is one I built some time ago for machining the bore of golf ball mortars.


Sorry the photo is blurred, but you can see the background pretty well.

That bar in the slot commects to the tailstock spindle for a positive feed.

You could also make the rotating part out of a gear and feed with a piece of rack. A worm gear is also possible to use.

The rotating element is machined with a radius on the far side so I can set the tool with an od micrometer.

Don't know if it's good enough for bearing fits, but it is an idea.


10-25-2006, 04:48 PM
weird, good material here, but doesn't address your original question - how do you cut a that internal sphere shape like with self aligning bearing? failing a cnc,seems to me you could do it with a boring head and the work mounted on a rotary table, set at an angle to the mill table. never done it, but I was trying to puzzle out the geometry. my head hurts now, picturing 3 d motion like this does that to me. you can do the same but cutting on the outside to make a sphere iirc. on second thought may that wouldn't work......

10-25-2006, 05:41 PM
This is about as simple as you could get for doing a concave radius, but it's obiously limited by its own body, especially for truly internal radii.



10-25-2006, 06:14 PM
And if you make an internal spherical hole how would you be planning on filling it?

It's not magic tooling & procedures you need it's the ability to see that not everything is made by cutting away material, forging/pressing/stamping can help to assemble what may afterwards seem to be a part made using the old black stick with white ends:D

10-25-2006, 08:58 PM
There is a company called "Eagle Rock" who makes both a internal and a external radius turning tool by the brand name of "Shaplane"

Travers tool sells them not cheap, they are a worm and wheel driven device.



kap pullen
10-25-2006, 10:10 PM
How about this one?

I believe wierd wants something that won't have
the pivot intersecting with the job like the others posted?


It has a worm drive and adjustable radius slide.


10-25-2006, 10:22 PM
Kap,that is almost exactly what I had in mind.The first set of seats I need to turn will be 5" od x 1.25" long with a 72mm spherical bore inside.These will accept a standard 72mm od ball bearing pillow block insert.I need to make four or five for a prototype,if the design is approved I can build a dedicated fixture for the turret lathe.

Filling the recess is a snap,look at a pillow block housing,two filling slots milled in 180* apart.

Okay,made my mind up,worm gear driven it will be,may get fancy and whip up an insert tip for the business end.

Thanks all.