In this post I will show how to build a lathe tool for turning spheres (balls) and other curved surfaces, and provide links to measured drawings.
Before starting this project, I made a complete set of design drawings. As the project progressed, I made minor changes and additions - these have been incorporated into the final set of drawings. Links to all drawings are provided at the end of this post.
The design goals for this project were as follows:
- Use a limited number of tools - primarily a lathe, drill press, and hand tools
- Use readily available materials
- Enable finished tool to mount to lathe without any lathe modifications other than removal of the compound slide
- Enable finished tool to be adjustable for convex or concave curves of varying size
Not surprisingly, I also thought of a number of possible improvements that might be made to the original design. I will note these as shown below:
[Possible improvement: These will be shown in brackets and red italics.]
Getting Started - the Main Plates
 |
| 1/4 X 4 X 12 steel plate |
The foundation for the tool is steel plate. I began with a piece of 4 inch X 12 inch steel plate (1/4 inch thick) purchased from the local big box store (in this case,
Menards, but most any good hardware store should have this).
 |
| Smoothing (filing) the plate end. |
The steel plate is pretty rough to begin with, so the first thing I did was file one end smooth and flat. This is primarily for cosmetic reasons, as this is not critical to the operation of the tool.
For appearance sake, I did my best to file the end as smooth, square, and flat as possible (with slightly chamfered edges).
With a little care, a clean square end can be achieved by simple filing. However, as noted earlier (if you make this yourself) don't worry if it's not perfect as it will have no impact at all using the completed tool.
 |
| Cutting the plate. |
Next, measure and cut two pieces 3 inches long. I cleaned up the cut edges as noted earlier. The other two edges were left as they were on the original piece of steel.
 |
| "Facing" the plate |
The plate surfaces were then "smoothed" by mounting the plate in a four jaw chuck on the lathe, and then facing the surface. I used an indexable carbide tool for this, although I am sure that an HHS tool would work as well (possibly better, as this is an interrupted cut, which can cause chipping with carbide tools).
The top plate was faced on both sides; the bottom plate was faced on the top surface only.
[Possible improvement #1: I'm not entirely sure that this facing operation is even necessary; I did it on the assumption that smooth mating surfaces would improve the rotation of the top plate, and a smooth top surface would ease adjustment of the tool radius. I now think facing of the top surface is probably not necessary. Also, as an alternative to facing on the lathe it might be possible to simply grind one surface of each plate on a lapping plate.]
 |
| Cutting hole for rotation collar. |
Once the plate surfaces are ready, a hole is cut in the bottom plate for the rotation coupling. The bottom plate is mounted in a four jaw chuck, and the hole is cut by first drilling, and then with a boring tool. Note that this is a stepped cut, and that the "bottom" of the plate is the unfinished side.
 |
| Finished hole in bottom plate. |
The finished hole in the bottom plate is shown above. You will, of course, want to finish the hole as close to the drawing dimensions as possible. Slight deviations are acceptable, however, as the rotation coupling is finished "to fit."
Rotation Coupling
 |
| Blank for rotation collar. |
From the remaining piece of the original steel plate, as additional piece is cut as shown, and the corners are cut off to create a rough octagonal piece.
This octagonal piece is then faced on both surfaces in the same way as the top and bottom plates. After facing is completed the piece is left mounted on the lathe and a center hole is drilled.
 |
| Octagonal blank and mandrel. |
A mandrel is made using a 1/4" bolt, a brass washer, and a drilled and tapped brass rod.
 |
| Blank mounted on mandrel. |
The rough piece mounted on the mandrel; the brass washer was used to prevent the bolt from marring the face.
 |
| Turning the rotation coupling |
Using the mandrel, the rough piece is mounted in the lathe and turned to appropriate dimensions.
As the rotation coupling turned, it is tested frequently against the finished bottom plate to ensure a firm but smooth fit.
A good fit is critical in order to eliminate vibration and wobbling when using the finished tool, so take as much time as necessary to get this right.
 |
| Test fitting the rotation coupling |
The completed rotation coupling and matching bottom plate are shown to the left. The coupling should rotation firmly and smoothly (with a bit of lubrication - I use lithium grease).
 |
| Bottom plate and rotation coupling |
The final step in completing the rotation coupling is drilling the holes to screw it to the top plate. The photo above shows the completed coupling as well as the completed cam locks. However, I recommend drilling the coupling holes
before adding the cam locks.
 |
| Bottom plate with cam locks and rotation coupling. |
The coupling holes are drilled by marking out the locations on the coupling. The coupling, top, and bottom plates are then clamped together in a "sandwich" and four pilot holes are drilled through both the coupling and the top plate. Use a center punch to punch a couple of witness marks in both the coupling and the top plate to show the original alignment, so the parts can be re-aligned later.
After drilling the pilot holes, disassemble the "sandwich" and finish drilling the holes to the required sizes. Countersink the holes in the coupling, and tap the holes in the top plate.
[Possible improvement #2: In retrospect, I think it may have been better to screw the rotation coupling to the top plate from the top, using socket-head screws. Allowance would have to be made for clearance of the top radius-adjusting screw of course, but I think there would still be room for four or six smaller screws on the top. This would simplify adjusting the "tightness" of the coupling. Also, this would probably be easier to do than getting the countersinking right.]
Cam Locks
 |
| Bolts (before threading/cutting) and one brass collar. |
The cam locks are quite simple. They are made from 1/4" bolts hack-sawed to length, and drilled brass rod. The bolts should have a 0.45" shank (un-threaded portion); I used bolts with a longer shank and added additional threads with a die, then cut the bolts to length.
 |
| Cam lock and hex bolt. |
Two of the brass rod pieces are drilled on center, the other two are drilled off center to enable the brass collar to press against the compound slide as it rotates.
I was originally a bit concerned that the cam locks might not hold firmly enough, but I found in practice that they work quite well. The short levers are a little bit difficult to press in place - I generally push them down with the flat of a wrench handle.
 |
| Bottom plate with cam locks |
Once locked in place the tool stays in position, with no tendency to wobble or slide.
Tool Post, Adjustment Screw, and Brackets
 |
| Marking post for cutting later. |
The tool post is fairly straightforward, but there are a few tips which may make it easier:
As you can see in the photo to the left, I marked the post cap
before cutting it off. I then drilled the two holes for tightening the cap (on a drill press) and then removed the cap with a cutoff tool. This ensures that the holes are perfectly aligned. When drilling the holes to depth, be sure to allow for the width of the cutoff tool. Also, the cap holes will have to be drilled a second time to allow clearance for the screws.
Once the cap is cut off, the tool can be removed from the lathe and placed on the assembled tool base (which has been mounted on the lathe) to determine the appropriate length needed to get the cutting tool at the proper height. My tool post turned out to be a few thousands too short, but I was able to remedy this easily by shimming the cutting tool.
[Possible improvement #3: It might be worthwhile to intentionally cut the tool post a bit short at the top, and then shim as necessary depending on the cutting tool being used. Ideally, the tool holder height would be adjustable in some way, but I have not yet come up with a simple way to do this (other than shims).]
 |
| Side View |
Once the tool post is cut to the proper height, it must be drilled and tapped for the brass adjustment screw. My initial design idea was to do this so that the tool post would sit firmly on the base, but still be able to slide back and forth to adjust the cut radius. However, I found that when I did it this way, the tool post wobbled sideways when cutting (even though it seemed firm when just sitting on the base). My solution to this was to make a new tool post, but this time to place the adjustment screw just a bit high. This way, after adjusting the cut radius with the brass adjustment screw, I could tighten down the brackets holding the adjustment screw and hold the tool post down to the base with considerable force; testing showed that this eliminated the wobble. The downside to this is that the entire tool assembly has to be removed from the lathe in order to loosen the bracket screws, and then removed again after adjustment to re-tighten them.
 |
| Drilling brackets on drill press |
The holes for the mounting brackets for the adjustment screw were drilled as shown to the left. Hole locations were marked on the brackets, and the brackets and screw were clamped together and drilled on the drill press. Each hole was drilled and tapped one at a time, so that a screw could be inserted to assist in the clamping. This is a rather laborious process involving a lot of assembly and dis-assembly, but I think it is worthwhile to ensure good alignment.
 |
| Tool post - side view |
The hole for the adjustment screw was drilled on a drill press. In retrospect it may have been better to drill it on the lathe, with the tool post mounted sideways in a four jaw chuck, to ensure that the hole is drilled absolutely square to the post. A drill press will do the job, however, with a bit of care and a reasonably good drill press vise.
If you look closely at the side view photo, you will see that the right hand mounting bracket does not sit flush with the top plate - this is intentional due to the "slightly high" location of the adjustment screw.
[Possible improvement #4: Instead of mounting the screw holding brackets with screws from below, use socket head screws mounted from above. This would enable the brackets to be loosened/tightened without removing the tool from the lathe. This would mean that amount of clearance needed for the bracket (as it swings under the lathe chuck) would increase, but this could be offset by other improvements with respect to the rotation coupling to be mentioned later.]
 |
| Adjusting screw, tool post, and mounting brackets |
The adjustment screw is made from 3/8"-16 coarse threaded brass rod, with the ends appropriately turned to fit the mounting brackets. I ran an appropriately sized die over the threads to clean them up. I put hex nuts on the threaded rod to prevent marring the threads, before mounting the rod in a three jaw chuck on the lathe to turn the ends. A slot was cut in one end to allow for adjustment with a screwdriver. The slot was cut using a Dremel cut-off wheel. The mounting brackets were cut from a brass 0.25" x 1" bar, with the cut ends cleaned up with a file.
[Possible improvement #5:Instead of cutting a screwdriver slot in the screw, drill and tap the end for a (cemented in place) socket head screw. This would enable the adjustment screw to be turned more easily using a hex wrench.]
Turning handle
 |
| Top view - Turning Handle with Ball End |
A handle for turning the top plate is mounted in one corner at at 45 degree angle. The bracket for the handle is also cut from a brass bar; it is mounted to the top plate with socket head screws.
 |
| Top view - closeup |
The photo to the left shows a closer view of the turning handle mounting.
Also note that the cutting tool is intentionally mounted off center; this enables the tool to cut closer to the mandrel.
[Possible improvement #6: Instead of using a cylindrical tool post, use a tool post with a square or rectangular profile. This might provide more stability, and would also be easier to machine. Also, I plan to replace the carbide cutting tool with a HSS cutting tool - I think this may provide a smoother cut on brass.]
Drawings, Instructions, and Bill of Materials
A complete set of drawings, instructions, and bill of materials for building this tool is available in PDF format at this link:
ball-turning-tool-by-kaje.pdf
The drawings were created using LibreOffice Draw. The editable drawing file is located at this link:
ball-turning-tool-by-kaje.odg
Acknowledgments
 |
| Smokey checks the fine print. |
Thanks to Smokey for his tireless checking of the paperwork.
