Square threads are sometimes made with slightly sloping flanks, that may be why you are having trouble.. If you have any slips, you might find the taper by checking the unworn part of the thread.

Is it not a worm? Driving a worm wheel?

OP quotes 5 TPI

regards

Make a gauge. Draw out the 29.5 degree angle on, for example, an A3 size piece of paper with lengths derived from trigonometry. Let us assume you can measure to a resolution of 1mm. Calculate how different the angle would be if you are 1mm wrong at that size (opposite side at A3 size is 237.6mm, so calculate for 237mm and 238mm). Assess whether that error is significant.

From the approx sizes you give in the first post 0.08" crest, 0.12" gap and 0.125" deep you have an angle of just over 18degreed (9.1deg each side) which could suggest it mates with a gear cut with a 20pa cutter.

Are the gear teeth cut straight across or at an angle making it a skew gear to match the helix angle of the screw. If it is cut at an angle don't take any measurements off the side of the gear wheel as they need to be taken at right angles to the teeth which may increase the angle if that is what you are taking your measurements from.

Lastly if the handle broke off is the rest of the screw still OK? if so think about grafting on a replacement rather than cutting again.

I would seriously consider remaking the cast bloc which supports the end of the screw so the more of the screw protrudes, then weld a stub/extension to the spindle, carefully machine the welded area to the original size, and fit the handle etc to the extension on the shaft. It would take away all of the anxiety that appears to present with identifying the correct thread form and the making of a suitable cutting tool etc. If threads are a problem, real or perceived, Nicholas simply look for an alternative……it's what engineers do.

Not wanting to sound obtuse, but it does not sound like you have the experience needed to cut a difficult and critical thread like this one. It is not a beginner project. You would be far, far better off to remachine the end of the existing screw so you can re-attach the handle.

This is usually simply done by mounting the screw in the lathe and machining the broken end flat and with a hole drilled and reamed in it so you can then turn up a replacement end part with a spigot sticking out that fits in that hole. You then fasten the two together by whatever methods suits, possibly Loctite, or silver soldering, or even careful TIG welding, or drilling and cross-pinning. This is much more achievable than cutting an obscure Acme-like 5TPI thread on a long thin screw to precision tolerances if you are not an advanced level lathist.

Edited By Hopper on 13/07/2022 11:59:47

Looks like a helical gear and worm, not easy to get the worm cut correctly to mesh fully. Interesting information here that I often used to use KHK Gears

No – don't try cutting a new screw with that great long, fancy-thread, high-lead!

Go with the extension-piece approach. It is transmitting only your hand-power, and there seems nothing wrong with the rest of the mechanism.

It's more than the technical problems of work and tool geometry, shown above.

Do you have a heavy-duty lathe actually designed to cope with such coarse threads, i.e. one having a 2 or at worst, 4 TPI lead-screw and change-wheels to match? Or the equipment for thread-milling?

Cutting a heavy thread coarser than the lead-screw, especially on a fairly light machine, is bad practice. It puts very unfair strains on the machine; even with the recommended "only if you really have to" practice of disconnecting the normal power drive and operating the lathe manually, "backwards" , from a handle on the lead-screw.

.

Use the method Hopper describes, though personally I'd not try welding or soldering, for such tasks. It can risk distorting the screw itself. If there is enough shaft projecting from the bearing you can alternatively use a tightly-fitting external sleeve, with pins. This is arguably stronger as you are not reducing the screw's cross-sectional area beyond perhaps a light skim for fit.

.

Following MadMike's suggestion, I would look at the bearing part carefully to establish if you need actually replace it. For preference, you need only trim it back or counterbore it a bit.

The deeper you go into modifying the machine to cope with repairing one part, the more awkward traps you risk having to dig yourself out of.

.

If you really need replace the whole caboodle, a limited range of such screw material is commercially available, but expensive! Most is Metric trapezoidal but I purchased a length of ACME stock from HPC Gears, for a new, short-form lead-screw for the early-pattern gearbox I have fitted to my Myford lathe.

…

A Warning Tale:

Big coarse thread-cutting is something I looked into yesterday, with my Myford; as I need be able to cut screws and nuts of 1/5" or 1/4" lead. I've 5tpi single-turn, square-thread taps; but the latter will probably need be of 2-start 1/8" pitch. Not something I've yet had the nerve to try!

Now, my ML7's gearbox is labelled as giving 8 to 56 tpi but in fact cuts half those indicated, so for Brass threads, (26 tpi) is set as showing 52tpi. Its finest is 28tpi (1/8" BS Pipe), and it can't cut 32 and 40 tpi.

That's due to the primary pinion on the headstock having 24 teeth, not the theoretical 12 teeth demanded but which would make the pinion too small… even if one is available!

So it will theoretically give 4 and 5 tpi pitches, but that is very cruel to a faithful little lathe even older than me. I thought I had actually broken it when experimenting with only a 0.005" cut on Whitworth form thread. Luckily I had just forced the banjo gears out of mesh, fortunately without broken teeth as they had stayed sufficiently engaged to jam.

Also though, I have a Harrison lathe with 1/4" lead screw, which will probably handle those threads. Without looking it up I do not know its recommended maximum cut pitch but I will still need treat it nice and gently; with tools formed on a tool-and-cutter grinder. (A Hemingway Kits, 'Worden'-design.)

Edited By Nigel Graham 2 on 13/07/2022 13:22:46

I have a lathe that uses a set of worms and wheels that fit into each other perfectly.

I think the gear wheel and worm in the first photo doesn't appear to be made to fit into each other properly, they would jam. I wonder if one or the other are an incorrect replacement from some time in the past

I have the same milling machine.

The broken leadscrew drives the x axis when the worm wheel rotation is locked.

The leadscrew also acts as a rack. When the worm wheel is unlocked it acts as a pinion, allowing rapid traverse with a handle.

On my list of jobs to do is to get rid of this drive as it is a source of vibration.