Transposing gears, the crazy way

[RandyWilson]

We won't get into why I bought an old American lathe when I had banished all things Imperial (and Whitworth) along with all things pushrod from my stable thirty years ago. But I did, a South Bend 14 1/2 dual tumbler. The only thing in my defense is the lathe came with all of the stud gears for the metric kit, only missing the 120/127 transposing gear. No problem? Problem.

I had read that the lathe used an odd 14dp change gear set. I verified that the stud gears were indeed 14dp. No one, not even Boston Gear, lists 14dp. So the hunt went on in odd spurts and starts, usually when I needed to cut a metric thread. 14DP is tough. I looked at the math of going to 127/100. It can be done, but would require a different stud gear set. So I looked into doing the 37/47. I determined the diameter was too small for the gears to engage. The 63/80 might work, but again, it would require making all new stud gears. In 14dp.

Then in a different thread, we learned how to make gears in FreeCAD. Alright, here we go. Drew up the 120/127 gears, and went searching for someone that could print them. It didn't take too long to find someone that was willing to print from my CAD files. They wanted to use their own slicer, but didn't want involved in the actual design. Hot damn. Yea HAH. So off to the design I go. I pulled the idler gear off of the lathe to measure the bushing diameter, hub depth, and all that needed stuffs. Bore is 7/8, hub depth is 1 5/8. Going good, nothing hard here. But something doesn't look right. Something is just... off. I take one of the 14dp stud gears and roll it around the idler. It does NOT mesh.

This is something I have not seen in any of my diggings, not here, not PM. The SB 13, 14.5 and 16 all use the same gearbox ratios, the came 6TPI lead screw pitch, and the same tooth count on the metric transposing set. What I have not seen anywhere is while the SB13 is 14DP as often quoted, the 14.5 is 12DP!! You 16 owners are on your own. I would *assume* they would be the same, but.... There is a clue in the SB parts manual. The shipping weight ont he metric kit for the 13 is 73lb, the 14.5 is 121, and for some reason the 16 is only 65.

So my change gear set is from a SB13. And I'm back to square one. At first I thought about making the transposing gear and the 64 tooth gearbox gear in 14dp and just switching out the entire set as needed. And I could make a 24 stud gear and a xx tooth idler so I can just leave it at DP14. But that's not the way SB intended life to be. And I really didn't buy all this machining equipment to have someone ELSE make things for me. And I'm skeptical of plastic drive gears keyed to a shaft. I have a DRO equipped mill. I have a lathe. I have a dividing head. I have a large rotary table. I even have aluminum blanks of approximately the correct size. HHS 12dp gear cutter sets are available from China.com.... ought to be good enough for low speed lathe gears in aluminum. The math is simple. I have zero experience. Checkity check, we're good to go.

Next up, teaching the dividing head a new number. I've looked around for a cheap 127 reference. Ironicly, the one odd source I found was a late MGB starter ring gear. But alas I banished those from my garage thirty years ago (see above). It's looking like I am going to end up drilling my own index plate. Damn prime numbers.

 

[silverhawk]

I am in that same boat. My dividing head uses a 4" index plate. I cannot have all 127 holes in the same circle because it won't fit. I will drill a 127 hole plate, then use that to drill another 127 hole plate, and repeat once more. It should then be accurate enough to make a gear.

Of course, you can always buy a 127 tooth gear and index based off of that. It might be easier, but not as cheap.

joe

 

[ttabbal]

Could one 3D print the first dividing plate to avoid drilling 127 holes in a throw-away plate? It would be decently accurate, so you might be able to skip the extra drilled plate. How are you getting the holes all in there? Multiple rows with an offset? Larger diameter?

 

[RandyWilson]

As I see it, I have two ways. First, I could build an arbor, adapters, supports and a lock to use a different gear, or I am sure I can borrow a MGB (127 10DP) flywheel from an old employer. This is a lot of one-off fabrication and dialing in. Or, I can start a new plate for my dividing head with the first pattern being 127.

I have said before that normal avoids me. Everything I get near ends up being odd, rare, weird and obscure. One morning about a year ago this ad popped up on CL for a dividing head, among other items. I talked to him and found he had bought the machine equipment from a farm auction and was reselling what he didn't want. I knew the general area of where he was, but Google maps wouldn't show it. When I dropped a pin at the location, Google maps would only say it was "near North America" and gave me hiking instructions. We shall say this guy was a bit back in the swamps near the Va NC line. His business was rebuilding Detroit Diesels. This dividing head was a big honking chunk of cast iron. And it didn't turn. It was a lot bigger than I was expecting. The thing is 100lbs, I was driving a little two seat convertible, and I was fresh off of abdomen surgery. I bought it or $50 as I remember, and the guy wouldn't even help me load it. I got it home, cleaned it up some, found a name. Kempsmith. Vintagemachinery even had a manual for it, dated 1916.

So, I have this antique 10.5" dividing head that came with only one of the two plates. The plate I have goes to 117 holes and is 7.25" in diameter. According to my measurements the thing will hold a plate with a circle up to 9.25". Indeed, the manual says optional plates are available up to 199 holes. So my thinking right now is to pick up a hunk of plate, cut out a circle around 8.5-9 inches, and drill the 127 pattern. This will leave me plenty of room, both in and outside the 127 circle, to add in anything needed that's not on the first plate.

 

[Lo-Fi]

Does your dividing head have an auxiliary input to the plate and output from the spindle?
Mine is an antique and has these features which allow differential dividing by gearing the spindle back to the plate.
If so, we can figure out what gears you'd need to cut to get the gearing right (hopefully with the plates you have) to index what you actually want to end up with. I've been through the journey myself and had a great time cutting gears - mine only came with a single, lonely 30 tooth gear.
Don't worry about slogging through charts, they're never complete. I've written a pc program that runs calcs based on what you actually have that you're welcome to use if you have that option and decide to go that route, of course. You did say "normal" avoids you

Also, for what it's worth: 3D printing dividing plates works very well too. I've printed myself a plate with 8 different patterns on it and successfully cut gears with it. It won't last forever, but it's plenty accurate. May be a good option, even if you farm that out.

 

[craptain]

Depending on the dro on your mill, it should be possible to program a 127 hole "bolt circle. Perhaps not conventional but....

Sent from my ONEPLUS A6013 using Tapatalk

 

[benmychree]

I was faced with the same problem, but had a universal dividing head and gears, the gears are specified for 127 divisions in the B&S book on milling, titled "Practical Treatise on Milling and Milling Machines". The gears needed to be cut, the 120 and 127 tooth were too big to cut on the dividing head without raising blocks, so I cut a smaller pitch gear to fit my antique Flather automatic gear cutter, and cut the full size coarser pitch gear on it. I can now cut 127 tooth change gears for others if they make the blanks, and have all the cutters for all even pitches and some odd numbered pitches, and some metric module pitches as well.

 

[benmychree]

Depending on the dro on your mill, it should be possible to program a 127 hole "bolt circle. Perhaps not conventional but....

Sent from my ONEPLUS A6013 using Tapatalk

Likely, you would have to make a longer sector arm to reach the larger circle, and some dividing heads will not accomodate larger plates

 

[RandyWilson]

I may or may not have to block up the dividing head. I need to get it up on the table and measure. It's a nominal 10 1/2" swing, the gear maths to 10.58" diameter. I most likely will have to block it with something like an 1/8" parallel. It will be that close.

In my case, the sector arm will handle the factory 9 7/8" optional discs. The sector won't, but mine are "broken" right now anyway.

The DRO has already been taken for a test pass at 127. That and making sure the divider would handle the disc size was the first step.

 

[silverhawk]

I think both options should work. If you can use the bolt circle function on the dro, that will make it easier. Use that to make a new index plate.

If you can't get the dividing head onto the mill, though, that's a problem. In that case, the 127 tooth MGB flywheel will work if you can key your blank to the flywheel and lock them together. Then just lock a vertical shaft to the mill table and a way to index off of the flywheel. You would need an adapter for the flywheel-to-blank. It should work well enough.

joe