understanding lathe change gears

Hello

This is a very noob question…. I’m intending to buy a lathe which I will eventually want to cut 26tpi threads (for vintage british motorcycles) and I’d like to understand change gears and thread cutting a bit better before buying.

In my budget and location (New Zealand) there are limited choices and most have a metric leadscrews. I’m eyeballing the optimum Tu-2506v, but it doesn’t specify 26tpi. I’ve read that an imperial leadscrew will cut imperial threads more accurately than a metric leadscrew will. But many with metric leadscrews also specify combinations for imperial threads (like the tu-2506). I assume it’s about maths and what % error I’m happy to live with, and not so critical when the threaded item is not too long. So…

1. is it true that if I can calculate the tpi or pitch from the given change gears then the machine will cut it, whether metric or imperial leadscrew? (assuming the gears will physically fit on to the banjo)

2. if so, what is a ‘reasonable’ % error on the pitch that I would want to live with? I’ve calculated the error at up to 0.3% from the change wheel combinations shown on the lathe I’m looking at (metric leadscrew-->imperial thread).

3. Is it thereby true if I buy (eg off ebay) some other change gears of other teeth number then I’ll get even more options? Any ‘gotchas’ on this?

Thanks for any help and clarify you can offer

Mike

Mike,

Not familiar with that particular lathe, but it looks like a modern version of mine. (Pics in an album.) If it's any help, my lathe has a 3mm pitch leadscrew and the chart supplied with it shows, amongst many others, a gear train to cut 26tpi. That is: spindle gear - 50 tooth, driving a 65 tooth which is coupled to an 80 tooth, driving a 42 tooth gear on the leadscrew. Others will be along soon to confirm or dispute this, no doubt! There may well be other combinations of gears, but this is what I can achieve with the gears supplied with the lathe as standard.

John

Thanks John. The TU-2506 has a 40 tooth spindle gear and 3mm pitch leadscrew. I can calculate combos that produce 26tpi (+/- .05%) so am assuming it will work. just not sure if I'm right and I'm left wondering if it is just maths why don't the manufacturers specify more thread possibilities on the label. maybe there is more to it?

1. True

2. This depends a lot on the length of the engaged thread, I'm sure you have screwed one screw into another hole and found the screw goes in so far before locking up when the two are different threads. Same applies with an error, a nut that may only have 4 or 5 turns of thread in it could work with a larger error than say a adjuster rod with 20-30 engaged turns.

3. If there is a mathmatical solution with other gears than those supplied then yes provided they will fit the banjo you can buy and modify them (bore & keyway) or get them 3D printed.

A quick look at the table on the front of the machine in question I can get 0.006mm error with stock gears

40T spindle driving 50T, that 50T driving 85T keyed to an 55T driving a 80T on the screw.

thanks Jason.

3D printing is an interesting idea

also for the 40T spindle driving 30T, that 30T driving 65T keyed to an 45T driving a 85T on the screw then I get 0.0005mm error per turn, or 0.046%.... yes?

so i conclude... it can be done!

so why don't they print these threads on the label?

Hi Mike,

I had the lathe you mention, in my opinion the low speed is too fast for screw cutting, from memory its around the 150 rpm mark.

I also found the electronic speed control was rubbish so bad in fact I replaced the motor and controller with a 3 phase vfd system.

I believe the non variable speed version of this lathe has a much lower bottom speed.

The other factor is that in NZ anyway there are no spares or accessories to be had for these lathes.

Sorry to put a damper on you but its better to know these things before you buy.

thanks for that info -eek, not sounding good for the 2506v, tho better to know while the money is in my wallet

Edited By Mike B 1 on 09/12/2018 09:17:00

Because the label would get very big, very quickly!

If you check out Brian Wood's book, you'll see the options for any one lathe can easily fill a few pages.

If it will cut that thread it should be on the label, other producers do it.

What you will find is there can be a lot of common thread pitches missing on asian imports and no change wheels to suit off the shelf.

I'm sorry to resurrect this post, but I had a lot of difficulty in figuring out the maths involved and the purpose of the 40T gear that Mike and Jason both refer to.

I had even more difficulty with John Hinkley's input and especially with the chart he has on his machine. The only way I could get the pitch mathematics right was to assume there is a hidden reduction gear supplying the so called spindle gear Z1. In John's case that is as high as 1/4.5 which could be supplied by 20 driving 90 but hidden within the casing before it emerges to carry Z1

I am used to seeing all the working gearing of a change wheel set up exposed to the eye and in the absence of any direct [ie hands on] knowledge of these Asian import lathes, am I correct in my assumptions?

Brian

Brian, the far eastern machines tend to have a fixed gear on the spindle in this case the 40T I mention. they also have a limited gearbox that will give 2:1, 1:1 and 1:2 so that same setup for 26tpi coul dalso be used to give 13 and 56tpi without having to alter whats on the banjo.

Worth looking at the front of the Optimum machine that was being discussed as they have put a very good graphic on the front rather then the usual charts that a lot of people can't follow, scroll along to the 7th photo, A, B & C are the 3 gearbox positions.

I could not work out John's gear train.

Pic of the 40T gear on my machine with the train below

Edited By JasonB on 10/12/2018 15:13:14

Jason.

Thank you for that explanation, you have told me this in the past, but I had forgotten. I'll look up that model as well for my education.

In John Hinkley's case, I checked through a selected number of pitches, both metric and imperial and I found the same degree of mismatch in both. Just using the figures straight from his chart and multiplying by the leadscrew pitch in the normal way [John quotes 3 mm] you will find that all the pitch results are too coarse by a factor of 4.5 times.

Thus, include that in the calculation as a preliminary reduction before Z1 [ 40/180 would allow for a sensible bore size on the lathe spindle] gives results which at last make sense. It is though, in my view, a very clumsy way of going about things, first of all gearing down and then gearing up again, certainly from 1/2 leadscrew pitch and coarser which will put a lot of loading on the gear teeth, some are quite low in tooth number as well.

The other thing that puzzled me was the appearance of a 63 tooth gear in the change wheel set up for a metric result on a metric leadscrew lathe. I can understand it being used for conversion to imperial pitches, but not that way. The maths works but they could be so much simpler.

Do these lathes not have a back gear set up of any sort to give screwcutting a better chance?

Regards

Brian

Brian, this may throw some light onto John's lathe that is nothing like the Opti machines or our Warco/Chester etc.

Looks like the belt from the motor drives the input to the 6 speed spindle gearbox and there is a shaft that comes out of that presumably geared up or down with Z1 on the end and then drives the rest of the train, that must be where the missing 4.5 is hidden.

There is no backger but unlike a Myford they don't have screw on chucks so you can happily cut with the machine in reverse, tool at the back and at a much faster speed so you get the torque that can be lacking at slower speeds. the added bonus is that like this they cut away from the chuck or a shoulder so you don't have to panic about stopping quickly at the end of the cut.

My imperial machine has a 63t in the gear set to get closer when cutting metric, usual supplied gears are 25 through to 80 in 5T steps plus an extra 80T for when being used for fine feeds.

Looks like an extrac chapter will be needed in your book when it comes round to reprint time

Brian, JasonB and Mike B1,

Jason is spot on with his assessment. I should have checked more carefully when comparing my lathe with Mike's. Outwardly similar, internally different. There is, indeed a geared headstock, the output of which is NOT the spindle, but as Jason correctly surmises, a separate layshaft, driven at a constant ration to the spindle. Hence the discrepancy in the gears. In my defence, I haven't been able to get near the lathe for weeks, let alone open it up due to the new kitchen cabinets and appliances cluttering the garage awaiting fitment, nevertheless I ought to learn not to jump straight in and land in my mouth with both feet, didn't I?

So ... Ignore my input in its entirety, it's a load of old hogwash in this instance!

Think I'll take up knitting fog - it's easier.

John

Hello John, Jason and Mike B

I don't think you need to make any apologies John. I was trying to make sense of the chart out of curiosity really, it was the appearance of 63 tooth gears in the metric threading that started me thinking as that was something I hadn't seen before.

So, I have learned something today from you both and thank you, but whether I have the stamina to put together another chapter for the book is a rather debatable point.

The mauling it got going though type setting with a non technical person doing the job was quite enough for one experience! My wife, bless her, and I had 7000 individual entries to cross check in the tables alone. It took 4 proof readings before I felt we had got there, but there are still typos we all missed

Regards

Brian

Edited By Brian Wood on 10/12/2018 17:55:27