What's new
What's new

I made a 62.5 tooth gear. Help me fix this

VanillahGarillah

Aluminum
Joined
Oct 10, 2022
Location
North of Baltimore
Purchased 1940s 10L that was originally procured by the Army and then transferred to the Navy. I had a problem with the Apron that caused the lathe to lock up every once in a while. I was quick on the draw to shut things down, until I wasn’t and I ended up modifying a 66 tooth gear into a 62.5 tooth gear.

Removed the problem gear to figure out what I needed to make, and started comparing my unusual mix of gears compared to what the parts diagram shows I should have.

Please help me figure out what the Navy was thinking with this gear setup. My goal is to cut a new gear so I can thread a chuck backplate to 2 1/4-8. With this odd gear setup, will the lathe cut threads according to the typical single-tumbler setup? And if not, what mix of gears do I need to make it work like it was originally designed?

Thanks for any help.

James
 

Attachments

  • IMG_1625.jpeg
    4.5 MB · Views: 107
Nice photo and clean labels!
So, the 40/24T cluster spins at the same rate as the spindle, either same or opposite directions as determined by the 28T tumblers.
The 24:64T reduction is 3:8, and the 22:66T reduction is 1:3 for a total reduction of 1:8.
I'm not familiar with either your machine or its standard gearing, but your labels suggest the damaged 66T is part of a 66/24T cluster. If you stack a 24:66T reduction (the 66T here being the wide-face red-painted gear driving the feed) on top of the 1:8, the result is 1:22. Does that match the "typical single-tumbler setup"?
It's quite possible I'm misinterpreting your intentions, as the reduction stages in the photo are not in mesh. As shown, there would be no feed at all.

Please list the actual gear train you intend to use, starting with the 40/24T cluster, and tell us what the lead on your lead screw is. (8TPI?) And for those of us not familiar with old Southbends, is there a threading gearbox between the red-painted gear and the actual lead screw? List each cluster gear as driven/driver. The number of teeth on a pure idler doesn't matter, so you can just write "idler".

Assuming your lead screw is 8TPI and no threading gearbox, to cut 8TPI threads you want the lead screw to turn at the same speed as the spindle. If there is a gearbox, you'll have to determine what SB wanted the input reduction to be, presumably from a diagram in your manual.
 
The gears are not in mesh because I removed everything to get the problem gear off and then just put it back together for the purpose of marking the gears and taking the photo. I wasn’t going to press the gears back in place only to have to press them out again.

I guess my question is what is/was the standard OEM gear setup for this lathe? I’ve seen parts diagrams and parts lists with various toothed gears, but I haven’t found the 66T or 64T gears listed.

The lathe has the standard (single) quick change gear tumbler that doesn’t look like it’s been modified at all, and the lead screw is 8TPI.

My question boils down to this: If I make a replacement 66T gear, can I expect the lathe to cut threads according to the chart on the quick change tumbler. Probably an easy question for South Bend 10L aficionados. I would also like to know the typical gear tooth count for the 10L lathe, because mine seems to be odd.

James
 
Purchased 1940s 10L that was originally procured by the Army and then transferred to the Navy. I had a problem with the Apron that caused the lathe to lock up every once in a while. I was quick on the draw to shut things down, until I wasn’t and I ended up modifying a 66 tooth gear into a 62.5 tooth gear.

Removed the problem gear to figure out what I needed to make, and started comparing my unusual mix of gears compared to what the parts diagram shows I should have.

Please help me figure out what the Navy was thinking with this gear setup. My goal is to cut a new gear so I can thread a chuck backplate to 2 1/4-8. With this odd gear setup, will the lathe cut threads according to the typical single-tumbler setup? And if not, what mix of gears do I need to make it work like it was originally designed?

Thanks for any help.

James
Just make a mating 18.5 tooth gear to the left, and another 62.5T to the right, and all is good...Seriously, several videos are out there brazing the broken teeth and re-machining them. If I recall, Keith Rucker has one, among others. You may want to reinforce them with some pins.

 
Just make a mating 18.5 tooth gear to the left, and another 62.5T to the right, and all is good...Seriously, several videos are out there brazing the broken teeth and re-machining them. If I recall, Keith Rucker has one, among others. You may want to reinforce them with some pins.

I watched his video several times and thought about it. I would have to repair 4 teeth in a row that sheared off. Getting the alignment and tooth shape to match would be a big problem.

I decided to get into gear cutting for a while and replace all the gears with new ones that mesh better with less chatter.
 
Get a 66 tooth gear and turn the middle out to make it a ring gear. Turn the OD off that gear and make it a close fit to the ring gear. Now fix the ring gear on with loctite and carry on.
 
I watched his video several times and thought about it. I would have to repair 4 teeth in a row that sheared off. Getting the alignment and tooth shape to match would be a big problem.

I decided to get into gear cutting for a while and replace all the gears with new ones that mesh better with less chatter.
Yeah, alignment could be a problem, but balance time to braze and fix versus making a new one from scratch. This is not a transmission for a 1000hp 300-MHP car :). That gear is pretty far down the gear chain and probably never turns faster than a couple of hundred ripms given the lathe tops out around 1.6k.

Looking at the picture again, it looks like this and the other gears had repeated visits to the dentist :).
 
Yeah, alignment could be a problem, but balance time to braze and fix versus making a new one from scratch. This is not a transmission for a 1000hp 300-MHP car :). That gear is pretty far down the gear chain and probably never turns faster than a couple of hundred ripms given the lathe tops out around 1.6k.

Looking at the picture again, it looks like this and the other gears had repeated visits to the dentist :).
The Navy probably used this lathe to design the LCS. This explains a lot about that multi-billion dollar failure.
 
To answer the op, I believe your 66 broken tooth gear is correct. Btw, the 66 tooth broken gear is part of the "sliding gear" which you set in or out based on the threading chart, along with tumbler and shift lever on the single tumbler qcgb. During normal lathe operations I leaving it pulled out to keep leadscrew from screaming.

There are a variety of reasons to bust teeth. Really it could be anything from hard debris in gear teeth, to crashing the machine, to misalignment of qcgb/leadscrew through apron, which can cause binding and tension.

When re-assembled. Put reversing gears in neutral, drop tumbler handle so its not engaged, apron NOT engaged. If you grab sliding gear in the "out" position, you should be able to spin the end gears by hand with it. Then leave apron NOT engaged, but put the tumbler up. Grab sliding gear again and try spinning again. If your alignment is good you should be spinning it by hand and see the lead screw turn.

If its bound up or too tight, you need to get alignment from qcgb, leadscrew, and end bearing correct. I'd start by loosening the screws to hold up qcgb and end bearing, plus the adjuster on bottom side of qcgb. The goal being to get leadscrew dead center of worm gear inside apron, for nice clean movement.

Gears that I know are different teeth number or ratio, based on lathe size are: twin gears(reversing gears) and the stud gear. As far as the rest, I believe they maintain the same ratio on the end on 10", 13", 14.5 and 16" lathes. As an example this is how my 16" is (I'm also set up with factory electric fwd and rev) only 3 gears with different tooth count:
582.JPG

My parts book only shows number of gear teeth for stud gears. However, only for twin gears(reversing gears), can you tell number of gear teeth by the end of part number. My pic may be hard to read, but for 10", twin gear part numbers have 28 at the end, where 16" show 32 & 36 which match my 16" gear teeth number:
581.JPG

Off topic, but there are optional stud gears that can be installed to obtain 27 tpi, if you ever happen to need to cut threads 27tpi. The pic shows part number plus chart for qcgb locations for that:
580.JPG

You mentioned cutting your own backplate. That may be a goal for you, not sure. But you can buy semi machined back plates. Spindle thread is cut. You sort the rest to get the chuck fitted. Search ebay: 2 1/4 × 8 chuck backplate, and look for semi machined.
 
Last edited:
To answer the op, I believe your 66 broken tooth gear is correct. Btw, the 66 tooth broken gear is part of the "sliding gear" which you set in or out based on the threading chart, along with tumbler and shift lever on the single tumbler qcgb. During normal lathe operations I leaving it pulled out to keep leadscrew from screaming.

There are a variety of reasons to bust teeth. Really it could be anything from hard debris in gear teeth, to crashing the machine, to misalignment of qcgb/leadscrew through apron, which can cause binding and tension.

When re-assembled. Put reversing gears in neutral, drop tumbler handle so its not engaged, apron NOT engaged. If you grab sliding gear in the "out" position, you should be able to spin the end gears by hand with it. Then leave apron NOT engaged, but put put the tumbler up. Grab sliding gear again and try spinning again. If your alignment is good you should be spinning it by hand and see the lead screw turn.

If its bound up or too tight, you need to get alignment from qcgb, leadscrew, and end bearing correct. I'd start by loosening the screws to hold up qcgb and end bearing, plus the adjuster on bottom side of qcgb. The goal being to get leadscrew dead center of worm gear inside apron, for nice clean movement.

Gears that I know are different teeth number or ratio, based on lathe size are: twin gears(reversing gears) and the stud gear. As far as the rest, I believe they maintain the same ratio on the end on 10", 13", 14.5 and 16" lathes. As an example this is how my 16" is (I'm also set up with factory electric fwd and rev) only 3 gears with different tooth count:
View attachment 416768

My parts book only shows number of gear teeth for stud gears. However, only for twin gears(reversing gears), can you tell number of gear teeth by the end of part number. My pic may be hard to read, but for 10" but twin gear part numbers have 28 at the end, where 16" show 32 & 36 which match my 16" gear teeth number:
View attachment 416769

Off topic, but there are optional stud gears that can be installed to obtain 27 tpi, if you ever happen to need to cut threads 27tpi. The pic shows part number plus chart for qcgb locations for that:
View attachment 416770

You mentioned cutting your own backplate. That may be a goal for you, not sure. But you can buy semi machined back plates. Spindle thread is cut. You sort the rest to get the chuck fitted. Search ebay: 2 1/4 × 8 chuck backplate, and look for semi machined.
This is the info I was looking for, so thank your for that. I would promote you from Titanium to Diamond or Platinum or Benitoite (if that‘s a level). I didn't know that my gear set up was normal. Thank you.

As for my gear’s missing teeth, you hit the nail on the head with this:

”misalignment of qcgb/leadscrew through apron, which can cause binding and tension.”

My apron was the source of this problem (indirectly). I replaced the original leadscrew and the lack of wear on the new one may have removed some play in the apron worm gear. I know you know this but, the worm gear sits between it‘s two bearings and those bearings are held in place by the bearing ‘nuts’ for lack of a better term. Those nuts keep their place via two pins. There is a pin that passes through the nut and into the bearing but stops short of the inner bearing surface. When I would engage the power feed, it would work most of the time, but eventually it would crash because the worm gear would shift to the right and lock everything up. When I tore the lathe down, I stamped each bearing and its corresponding nut, but I didn't stamp which one was right or left, like an idiot. That’s what broke my gear. I have taken my apron off so many times that I built an apron stand. Anywho, I found the source of the problem. Unsurprisingly, it was me. When I put the apron back together, I installed the bearing/nut pairs on the 'wrong' side. I switched them just to see if that would fix the binding problem, and it did. Haven’t had any problems since then. 🤞🏼

I couldn't find a suitable (thick) threaded backplate blank for my 6" Buck adjust true chuck, so I ordered the thick 8" back plate from Grizzly and I'm to the point now that I want to thread it and put it on the spindle to finish the work.

Thanks again for the info your provided. I am a mechanically competent tinkering hobbyist, but my hair is the color that makes you forget certain things, like marking parts as left or right.
 
You may know, or have done so already. . .The nuts and bearings you mentioned for worm gear. There should be a felt in the short channel on face of bearing that nut touches. The felt should lead to bottom of apron to wick oil from there. The faces of both nuts act as a thrust bearing against face of bearing. Switching nuts is not too critical, however the side toward tail stock was probably wore considerably, while the nut/bearing on qcgb side of lathe would have near zero wear. That's because while cutting with the lathe, the apron is always driving and working hard toward the chuck, and not working driving away from chuck.

Anyway, I recommend putting the non-wore nut/bearing to its original locale, to keep worm gear at its centered position. Then adjust the wore nut/bearing side until you have .002"-.003" end thrust on worm gear, that's .002"-.003" moving worm gear back and forth. Once adjusted, now drill a hole through existing lock pin hole in nut, use a drill size a couple thou's smaller than locking pin, and hammer locking pin. You're only drilling through threads of worm gear by using existing hole in nut. That'll remove the slop, but allow oil clearance on both nut and bearing faces.

Now for leadscrew alignment. When aligning any shaft, think of a circle and a clock face. Most common practice is get centered between 3 & 9 oclock, and then between 12 oclock & 6 oclock, or vice versa. And most likely need to do so in increments, to get totally centered. There is an adjuster on bottom of qcgb, it may be a bolt with 2 screws to either side. That adjuster is good for 3 to 9 oclock adjustment. . .

But what about adjusting between 12 & 6 oclock ? In the case of South Bends we can shim qcgb and lead screw end bracket down. And its usually always shim down. The reason is, on an old lathe, bed ways wear a little, and saddle/carriage can wear a lot. When saddle wears, it drops closer to the floor, as a consequence, the apron also drops straight down with this wear. As apron drops over time, it puts downward pressure on leadscrew.

I'd go to the auto parts store and buy a couple/few sets of cheapo straight feeler blades, to use for tests and possibly as my shims. You can do alignment by feel of your hands, by spinning the 66 tooth gear on qcgb, and by rolling apron/saddle assembly completely up and down ways. Everything should roll, spin, and move freely by hand when its right.

Use feeler blades as shims between bed and qcgb, and between bed and leadscrew end bearing. Maybe start at .005" and work in .005" increments until you find your best sweet spot of spinning 66 tooth on qcgb (tumbler engaged) and hand cranking apron/saddle up and down ways.

FYI, you can also shim bed rack down too, and for the same reasons. If there is excessive slop moving handwheel on apron back and forth.
 
Last edited:
You may know, or have done so already. . .The nuts and bearings you mentioned for worm gear. There should be a felt in the short channel on face of bearing that nut touches. The felt should lead to bottom of apron to wick oil from there. The faces of both nuts act as a thrust bearing against face of bearing. Switching nuts is not too critical, however the side toward tail stock was probably wore considerably, while the nut/bearing on qcgb side of lathe would have near zero wear. That's because while cutting with the lathe, the apron is always driving and working hard toward the chuck, and not working driving away from chuck.

Anyway, I recommend putting the non-wore nut/bearing to its original locale, to keep worm gear at its centered position. Then adjust the wore nut/bearing side until you have .002"-.003" end thrust on worm gear, that's .002"-.003" moving worm gear back and forth. Once adjusted, now drill a hole through existing lock pin hole in nut, use a drill size a couple thou's smaller than locking pin, and hammer locking pin. You're only drilling through threads of worm gear by using existing hole in nut. That'll remove the slop, but allow oil clearance on both nut and bearing faces.

Now for leadscrew alignment. When aligning any shaft, think of a circle and a clock face. Most common practice is get centered between 3 & 9 oclock, and then between 12 oclock & 6 oclock, or vice versa. And most likely need to do so in increments, to get totally centered. There is an adjuster on bottom of qcgb, it may be a bolt with 2 screws to either side. That adjuster is good for 3 to 9 oclock adjustment. . .

But what about adjusting between 12 & 6 oclock ? In the case of South Bends we can shim qcgb and lead screw end bracket down. And its usually always shim down. The reason is, on an old lathe, bed ways wear a little, and saddle/carriage can wear a lot. When saddle wears, it drops closer to the floor, as a consequence, the apron also drops straight down with this wear. As apron drops over time, it puts downward pressure on leadscrew.

I'd go to the auto parts store and buy a couple/few sets of cheapo straight feeler blades, to use for tests and possibly as my shims. You can do alignment by feel of your hands, by spinning the 66 tooth gear on qcgb, and by rolling apron/saddle assembly completely up and down ways. Everything should roll, spin, and move freely by hand when its right.

Use feeler blades as shims between bed and qcgb, and between bed and leadscrew end bearing. Maybe start at .005" and work in .005" increments until you find your best sweet spot of spinning 66 tooth on qcgb (tumbler engaged) and hand cranking apron/saddle up and down ways.

FYI, you can also shim bed rack down too, and for the same reasons. If there is excessive slop moving handwheel on apron back and forth.
Wow. Great advice. When I tore the lathe down, I replaced all the felt, so the worm gear felt is fine. I know this because I’ve had the apron off the lathe about a dozen times trying to figure out why the worm gear kept jamming. It works perfectly now, so I don’t want to mess with that.
I have feeler gages and plenty of shim stock. I wasn’t aware that the leadscrew could be adjusted. I’ll get right on that. I never thought to shim the rack, and I never thought about shimming the QC gear box.

Once properly aligned, should the QC gear box shift smoothly? As it is now, I can change the gears and shift the top lever, but it’s far from smooth.

I would now promote you to unobtanium status for your advice.
 
Once properly aligned, should the QC gear box shift smoothly? As it is now, I can change the gears and shift the top lever, but it’s far from smooth.

The qcgb will probably never be smooth. It must shifted with lathe powered off, so we often bump gear teeth head to head when trying to engage. Sometimes we need to give it a quick bump on the power switch to get gears aligned.

In something like a manual transmission, the gears are rotating when we change gears, but there are clutches, synchronizers, and such to ease the engagement. Nothing like that in a qcgb, so we need to kinda bump and roll a little, to get gear teeth lined up.

Unless you're cutting a great many variety of thread pitches, you don't need to change qcgb too often. I pretty much leave pinned to the same location for general lathe work. For me and what I do, usually the 64 tpi setting on chart gives what I like, in that, the sliding gear is pulled out.

When you get fancier with speeds and feeds based on work material, tooling etc, you might do more math to get it perfect. But for general stuff, I just leave it pinned to what works for me most.
 
Last edited:
You can fix that gear yourself if you want. Old school method:

SF_gear01.jpg


SF_gear02.jpg


My version of the fractional tooth gear:

SF_gear14.jpg


Finally got it right!

SF_gear16.jpg
 
OK. That's pretty impressive, but I'm going to make my own gears. Someone mentioned the 10L gears were all DP18, but so far, the ones I'm trying to recreate are all DP16 based on the gear cutter profile in the gears' teeth. Does it matter if so make these gears with DP18 cutters or DP16 cutters? I know this has to be a stupid question, but that heavy 10 factory DP18 gear comment stuck with me.
 
I meant will two DP16 Gears with, say 28 teeth each, function differently than two similar gears cut with DP18 cutters. Allmof the same pressure angle. The only difference would be gear diameter, correct?
 








 
Back
Top