Metric thread on imperial lathe, without reverse.

[Vernonv]

It's been a long time since I cut a metric thread and I can't remember if I have even cut one on my current lathe (which doesn't have a reverse).

Anyway, my question is .... is it possible to cut a metric thread on an imperial lathe, if I disengage the half nuts and reposition the tool at the same location and start on the same number on the thread dial, for each pass?

[bob ward]

I guess so, but your carriage will need to be accurately repositioned each time, maybe use a stop. Do a trial run first if you are working on something irreplaceable.

[mahgnia]

It's been a long time since I cut a metric thread and I can't remember if I have even cut one on my current lathe (which doesn't have a reverse).

Anyway, my question is .... is it possible to cut a metric thread on an imperial lathe, if I disengage the half nuts and reposition the tool at the same location and start on the same number on the thread dial, for each pass?

Vernonv,

It is extremely unlikely that you will get the correct positioning on each pass due to the metric thread not being a multiple of the lead screw pitch, even using the thread dial, unless the pitch chart for your lathe gives you this option.

The are a few possible options:
1. Fit a reversing switch to your lathe motor (usually a fairly straightforward job)
2. Turn the lathe in reverse by hand using a crank temporarily connected to the motor, lathe spindle, or one of the drive gear shafts.
3. Install a metric lead screw and half nuts (this may present it's own set of new problems)

Andrew

[Vernonv]

Do a trial run first if you are working on something irreplaceable.Yes, I will certainly do that.



1. Fit a reversing switch to your lathe motor No can do (and I have tried) - the lathe has a clutch and it doesn't work correctly when run in reverse.


2. Turn the lathe in reverse by hand I was hoping to avoid that if possible, but may just have to.


3. Install a metric lead screw and half nutsI can't really justify the expense or time to do that.

Any other comments/thoughts anyone?

[Big Shed]

Vernon, I don't know which lathe you have and this may be way off beam, but on the 9x20 lathes a very popular modification is to fit a tumbler reverse.

Here is one of those mods, there a numbe rof others on the net:

Steve Bedair's 9x20 Lathe Site (http://bedair.org/Tumble/Tumble.html)

[Vernonv]

Thanks Fred ... I have an old Nuttall lathe, which has a leadscrew reverse lever already.

I just did a search on here and realised that I asked this very same question some time ago (2010) :B, but unfortunately I didn't post what the conclusion was. Although I seem to remember that I ended up opting for an imperial thread.

I might have to do some testing tonight and see what my options are in regard to the leadscrew reverse.

[Stustoys]

leadscrew reverse wont work unless it only have one engagement postion and is spinning 1:1 with the spindle(I think)

There is one way that I'm aware of(but have never tired)
You have a fixed starting point for the carriage with the halfnuts engaged. Mark the chuck and the leadscrew at TDC. Take a cut. disengage the half nuts, move the carriage back the the start point. Wait until the marks on the chuck and leadscrew line up on TDC and engage half nuts.(I think lol*)

Stuart

* sure "sounds" easy enough. Though how easy it ends up being might depend on the TPI you are cutting and the TPI of your leadscrew

[Metmachmad]

Please bear with me here, but I must be missing something regarding the pitch of metric threads. Maybe someone can explain it to me (and possibly others also.)
Say you have got a 10mm thread with 1.5 mm pitch you are wanting to cut on an imperial lathe, to convert it to Threads Per Inch (TPI), because thats the way imperial lathes cut threads, by my reckoning that would equate to 25.4mm (1 inch) divided by 1.5 mm pitch = 16.933333 TPI.
What imperial lathes allow one to cut 16.933333 TPI, certainly not mine. Please explain.

[Vernonv]

What imperial lathes allow one to cut 16.933333 TPI, certainly not mine. Please explain.If you have the right change gears you can certainly get close, if not exact.

[Stustoys]

A 127-120 gear gets you exact
63-60(I think thats the one used) gets you close

Stuart

[Metmachmad]

If you have the right change gears you can certainly get close, if not exact.
Are you suggesting that 17 TPI would be close enough?!?!.

[Vernonv]

Are you suggesting that 17 TPI would be close enough?!?!.Possibly, depending on the intended application - thread pitch error is a compounding problem i.e. on a short thread interface (thread contact length) it's less of an issue, whereas on a long thread interface it is more of an issue.

However, often you can get the right metric thread from an imperial lathe if you have the right change gears.

[franco]

Vernon,

Some time ago I made an expanding mandrel to fit the outer end of the headstock spindle hole so I could mount change gears on it for direct indexing with the headstock spindle. After I made a 37/47 metric conversion gear I thought of making a handle for the mandrel to wind the spindle backwards manually. This got boring very quickly, so as a temporary measure (several years ago!) I made a quick release coupling to fit the mandrel and use a 13 mm electric drill to reverse the spindle after disengaging the back gear - not the half nuts - so the drill does not have to back-drive the motor and countershaft. As a temporary measure this works OK for the small number of metric threads I cut on this lathe - must get around to making a more permanent solution - one day - maybe..........

Using the tumbler gears to reverse the leadscrew as suggested above to wind the carriage back will not work because the relationship between the spindle and leadscrew positions will be broken.

Frank

[matthew_g]

leadscrew reverse wont work unless it only have one engagement postion and is spinning 1:1 with the spindle(I think)

There is one way that I'm aware of(but have never tired)
You have a fixed starting point for the carriage with the halfnuts engaged. Mark the chuck and the leadscrew at TDC. Take a cut. disengage the half nuts, move the carriage back the the start point. Wait until the marks on the chuck and leadscrew line up on TDC and engage half nuts.(I think lol*)

Stuart

* sure "sounds" easy enough. Though how easy it ends up being might depend on the TPI you are cutting and the TPI of your leadscrew
Stustoys has hit the nail on the head for a quick easy fix for a one off.
I have done this exactly as Stu has said for a one off job, If you had to make hundreds of parts I would be looking for other alternatives, but this way does work, there is no extra expense to justify and no extra 2 or 3 days of modifying your lathe for a 2o min job.
Good luck mate.


Matt

[steamingbill]

Have a look at the calculator in this thread.

http://www.woodworkforums.com/f65/lathe-thread-cutting-change-gear-calculator-118398/

Its good at showing how close you can get to metric threads.

Ironic - I just realised that it was written by Vernon the chap who started this thread -

There are some other calculators and examples here Screwcutting Programs (http://www.lathes.co.uk/latheparts/page14.html) the text below is copied from that page.


EXAMPLE 2: (Imperial leadscrew, metric thread required)
Leadscrew Turns per inch : 8 (1/4 inch pitch)
No. of teeth on primary shaft gear: 48
No. of teeth on 1st stud driven gear: 50
No. of teeth on 1st stud driver gear: 36
No. of teeth on leadscrew gear (driven) : 44
TPI = 50/48 x 44/36 X 8 = 10.185 TPI
Convert to millimeters : Pitch in mm = 25.4 / TPI
25.4 / 10.185 = 2.494 mm pitch
This equates to a 0.25% error against a 2.5-mm pitch.

Bill

[Ueee]

I'm not 100% on this, but other than Stuarts suggestion you can do it by reversing the leadscrew with the tumble reverse. You must however mark a tooth/valley on each gear and make sure you only engage the tumble reverse on the marks, for the cut anyway. Again, the half nuts must remain engaged the whole time.

I don't know why more lathes are not made with a single tooth reversing dog clutch like the hendys and some Holbrooks have. Solves the whole metric/imp threading issue.

Ew

[Oldneweng]

Thanks Fred ... I have an old Nuttall lathe, which has a leadscrew reverse lever already.

I just did a search on here and realised that I asked this very same question some time ago (2010) :B, but unfortunately I didn't post what the conclusion was. Although I seem to remember that I ended up opting for an imperial thread.

I might have to do some testing tonight and see what my options are in regard to the leadscrew reverse.


No can do (and I have tried) - the lathe has a clutch and it doesn't work correctly when run in reverse.

I cannot help with your question but I am interested in the fact that your Nuttall clutch doesn't work properly in reverse. Mine has a reversing motor on it and the clutch works fine regardless of direction. Maybe a different clutch or something.

Dean

ps My CY has that thing Ewan described. Harty69 explained it to me in relation to his lathe last weekend and I realised that my lathe had the same system.


I don't know why more lathes are not made with a single tooth reversing dog clutch like the hendys and some Holbrooks have. Solves the whole metric/imp threading issue.

[malb]

If you reverse the tumbler gears, you are effectively going to cut a left handed thread, it might or might not work by fully retracting the tool from the thread prior to running back with the tumbler reverse, then feeding back to the previous cut and adding the next cut before making the next pass.

The issue that will determine whether or not it works is whether you can change the tumber and have the gears engage repeatedly without needing to rotate the spindle or leadscrew even minutely. The few lathes I have dealt with (Southbend and Chinese) tended to have a tooth offset of about 1/5 to 1/3 tooth when operating the tumbler mechanism, meaning that the spindle needed a small tweak to engage the gears during the tumbler reverse operation. Remember that the tumbler mechanism is intended to select RH or LH thread before starting to cut the thread, rather than wind back the saddle in sync with the thread in the workpiece at the end of each cut.

My dad would have cut 1000's of threads with his SB in the 45 years he ran it, no chaser or dial gauge, always half nuts engaged continuously, cut lowest speed backgear or hand cranked, and returned by hand cranking at the chuck or faceplate.

[Michael G]

A while ago there was an article in MEW about this and the author made a threadng dial for metric threads on an imperial lathe. Because of the conversion between the two I think it had 127 teeth on it. This was because the first integer that converted inch to metric was 127.
For an occasional thread keep the half nuts in and reverse the spindle after winding out. Some people have made cranks or wheels that lock into the back of the spindle to make it easier to turn the spindle back.

Michael

[Vernonv]

I cannot help with your question but I am interested in the fact that your Nuttall clutch doesn't work properly in reverse. Mine has a reversing motor on it and the clutch works fine regardless of direction. Maybe a different clutch or something.Hi Dean. I did the test in 2010 and I found (if I remember right) that the clutch permanently engaged when I ran it in reverse. My Nuttall is a 1948, threaded nose model.

I didn't get a chance to do any testing last night, although I did get it mostly setup. Hopefully tonight I will have a go at it.

[PDW]

ps My CY has that thing Ewan described. Harty69 explained it to me in relation to his lathe last weekend and I realised that my lathe had the same system.

It solves more than the metric threading issue. You can screw-cut at higher speeds without risking a crash or needing lightning reflexes and you can use it under power feed to a shoulder and walk away to do something else without a worry about getting back before a crash. I use it all the time on my Monarch CY.

PDW

[Oldneweng]

Hi Dean. I did the test in 2010 and I found (if I remember right) that the clutch permanently engaged when I ran it in reverse. My Nuttall is a 1948, threaded nose model.

I didn't get a chance to do any testing last night, although I did get it mostly setup. Hopefully tonight I will have a go at it.

Mine is about the same period and chuck mounting is also screw on, as in threaded spindle. My lathe has a combo clutch and brake although the brake has stopped working. Needs adjusting I think. I had forgotten it even existed until I started to think about how the clutch operates. If the handle is not engaged the brake should be automatically applied. The brake consists of a friction disk that presses against the outer end of the clutch housing which has a raised disk for this purpose. Is this similar to yours?

291678

I have found a picture of yours and although it looks more pretty it is the same.

Dean

[Retromilling]

If you have the correct gear combination for the pitch then it is possible to disengage the half nuts without using a thread dial . You have to close the half nuts and pull the lathe forward to start near the thread start . Then put three witness marks on the lathe. One on the backing plate to head stock. One on the lead screw to gearbox housing and one on the bed way at the edge of the carriage .
Start the lathe and take the first cut and disengage as normal . Then move the carriage back to that mark , pull the lathe around until the other two marks line up if you miss go back and pull forward again to stop backlash , move the carriage in towards the chuck to line up and close the half nuts and you should be back in the same starting point. Increase the cut and start the lathe . That's as best I can remember I have never done it myself but I have seen it done by a person I knew years ago and on a you tube video a while back which I will try to find again .
I think this is the same one https://www.youtube.com/watch?v=FNajMA4I20A

[cba_melbourne]

It's been a long time since I cut a metric thread and I can't remember if I have even cut one on my current lathe (which doesn't have a reverse).

Anyway, my question is .... is it possible to cut a metric thread on an imperial lathe, if I disengage the half nuts and reposition the tool at the same location and start on the same number on the thread dial, for each pass?


Vernonv, the answer to your question is an unconditional no. And I am 100% confident about this.


If your lathe had a metric leadscrew, then you could open and close the halfnuts wherever you want and how often as you like and without using a thread dial, for those metric thread pitches that are contained as a whole number in your leadscrew pitch. For all other metric pitches you could use a thread dial if available. But since your lathe is imperial (eg has an imperial leadscrew), there is absolutely NO way to disengage the half nuts without ruining the thread being cut. Some people may claim you can with some tricks, but your chances of getting onto a "false match" are many times greater than hitting by chance the one and only engagement point that exists. Without a reverse on the spindle, your only option is as suggested earlier, leave the halfnuts closed and turn the spindle backwards by hand. Chris

[Vernonv]

Is this similar to yours?Yes, as you said, it's the same ... although I didn't realise it braked as well - I guess that doesn't work on mine either.

So your clutch works regardless of the direction of rotation?
Maybe I need to have a closer look at the clutch mechanism.

You don't happen to have a manual do you?

[Oldneweng]

Yes, as you said, it's the same ... although I didn't realise it braked as well - I guess that doesn't work on mine either.

So your clutch works regardless of the direction of rotation?
Maybe I need to have a closer look at the clutch mechanism.

You don't happen to have a manual do you?

Unfortunately I don't have a manual. Just had a quick look again on the net. I had a look when I first got it but had no success. I found a reference on PM from 2008, to a very worn Nuttall purchased by "RC99" whose location was given as "near Rockhampton, Queensland, Australia". Also a thread on this forum from 2009 with discussion from .RC. and an "Old Croc" who it appears was about to receive a manual. Title "Nuttall Lathe Manual".

I also found this picture of a very well preserved unit.

291783

This was on Lathes.co.uk. They don't have a manual.

I cannot help much but if you need pictures just let me know. They may be of some use. I broke the clutch housing by dropping the lathe when moving it into my shed and had to weld up a replacement. The clutch/brake mechanism is self supported on the shaft so the only thing the housing did was to provide a fixed surface to brake against and a cover over the top. This cover survived.

Dean

[Retromilling]

I'm not 100% on this, but other than Stuarts suggestion you can do it by reversing the leadscrew with the tumble reverse. You must however mark a tooth/valley on each gear and make sure you only engage the tumble reverse on the marks, for the cut anyway. Again, the half nuts must remain engaged the whole time.

I don't know why more lathes are not made with a single tooth reversing dog clutch like the hendys and some Holbrooks have. Solves the whole metric/imp threading issue.

Ew

It's not such an issue on a modern lathe with an instant stop foot brake . I cut metric and Imperial on an Imperial lead screw and never use a thread dial on Imperial . I just hit the foot brake and the lathe stops dead on , back out and reverse . On an instant reverse three phase lathe you can just whack the lever into reverse as you back out or use the brake . Not having reverse would be a bitch .

[Vernonv]

Ok I cut the thread. A few things to note:

1. My lathe has a brake :doh::U. Works really well at lower speeds (stops it dead) and slows the chuck much quicker than normal, at high speeds.

2. Reverse lever (for the carriage only) does not bring you back to the correct spot when metric threading.

3. I had to reverse the chuck by hand, with everything engaged to get back to the start of the thread. Of note was that I could change the gearbox from slow to fast to allow me to easily reverse the chuck and it reversed to the correct start position. I could then change back to slow to cut the thread.

4. I needed to cut an M16 x 1.5 thread (16.9333 TPI) and using the standard change gears (only 3 of them :no:), I could get (using the Lathe Gears program to work it out) to 1.511mm (16.8 TPI) which is a 0.79365% error. That was close enough for my application. :)

Only thing left now is to revisit the clutch operation when the motor runs in reverse - it's a 3 phase motor so will be easy.

[Oldneweng]

Ok I cut the thread. A few things to note:

1. My lathe has a brake :doh::U. Works really well at lower speeds (stops it dead) and slows the chuck much quicker than normal, at high speeds.

2. Reverse lever (for the carriage only) does not bring you back to the correct spot when metric threading.

3. I had to reverse the chuck by hand, with everything engaged to get back to the start of the thread. Of note was that I could change the gearbox from slow to fast to allow me to easily reverse the chuck and it reversed to the correct start position. I could then change back to slow to cut the thread.

4. I needed to cut an M16 x 1.5 thread (16.9333 TPI) and using the standard change gears (only 3 of them :no:), I could get (using the Lathe Gears program to work it out) to 1.511mm (16.8 TPI) which is a 0.79365% error. That was close enough for my application. :)

Only thing left now is to revisit the clutch operation when the motor runs in reverse - it's a 3 phase motor so will be easy.

I guess that you don't have the metric thread cutting gears for your lathe. They came with mine. I am going to have another look at my clutch/brake setup and take some pics. I would like to fix the brake but as the lathe is not really mine anymore I don't plan on spending much time or money. Might be a good idea to have some photos for when it is gone. Wont be far away tho.

Dean

[Stustoys]

Vernonv, the answer to your question is an unconditional no. And I am 100% confident about this.

100% is a lot.:D

I think I've just done the maths on cutting a 1.5mm pitch on an 8 tpi leadscrew.
Now if I'm right you get exact pick up every 127 pitches(60 turns on the leadscrew). Only 22 of the other turns will the leadscrew be within 36 degrees of TDC with the chuck is at TDC. So granted....... it could be tricky on the fly and at 75rpm... slow.

But this idea might work.
Make up a gage that allows you to move the carriage 7.5" +/-1/32(you just need to be on the correct "turn" of the leadscrew"
Cut your thread, turn the lathe off, open half nuts, move the carriage back 7.5" close halfnuts, cut thread, repeat.

I cant check my maths on the lathe as its metric.

IF the above idea works, another idea you could make up a thread chaser dial with a 60 tooth* gear, wouldn't that let you pick up on the fly?

Granted this wont exactly be fast as you'll effectively be cutting a 7.5" long thread regardless of how much you want.(though you do get quick return unlike full reversal, if the chaser works you wouldn't even be turning the lathe off)

Stuart

p.s. Surely I'm missing something?

0.5 pitch = 2.5" reset
0.25 pitch= 1.25" reset

0.5 pitch = 20 tooth gear
0.25 pitch = 10 tooth gear

p.p.s I've got to be missing something!

[Vernonv]

I guess that you don't have the metric thread cutting gears for your lathe. They came with mine. I am going to have another look at my clutch/brake setup and take some pics. I would like to fix the brake but as the lathe is not really mine anymore I don't plan on spending much time or money. Might be a good idea to have some photos for when it is gone. Wont be far away tho.No unfortunately the only change gears I got were the standard imperial ones - 30 (spindle), 72 (idler) and 120 (quick change gearbox). I can swap the 72 and the 120 around to give me a few more options. I would love a decent set of change gears, but it's cost prohibitive.

What change gears did you get with yours?

I took a photo of my clutch brake setup, so will post when I have my phone handy.

[Ueee]

What gears are needed Vernon/Dean? And what is the dp?

Ew

[Vernonv]

What gears are needed Vernon/Dean?Not exactly sure, as I don't know what was available originally, but I guess a 127 would probably be a good starting point.


And what is the dp?Don't know.

[chambezio]

I am looking forward to see where this thread leads. My 100 (plus) year old metal lathe is German in origin but is set up with the change gears for cutting imperial threads. Because of its age there is no thread dial. So to cut any thread, after you instal the appropriate gears you have to reference the cutting with marks on the chuck and headstock as well as a mark on the lead screw. I cut my first thread 1" X 8TPI using the above method, yes slow and tedious but I got a result. By the way the information is coming in with this thread, I may be able to cut metric threads as well

[Vernonv]

OK, so I got around to testing the clutch with the motor in reverse, and low and behold it works fine :? ... I'm not sure what strange phenomenon I experienced last time I tried it. Now I just need to locate a change over switch.

Dean, here is a pic of my clutch/brake setup for reference. If you need any more pics/info, let me know.

291805

[Ueee]

100% is a lot.:D

I think I've just done the maths on cutting a 1.5mm pitch on an 8 tpi leadscrew.
Now if I'm right you get exact pick up every 127 pitches(60 turns on the leadscrew). Only 22 of the other turns will the leadscrew be within 36 degrees of TDC with the chuck is at TDC. So granted....... it could be tricky on the fly and at 75rpm... slow.

But this idea might work.
Make up a gage that allows you to move the carriage 7.5" +/-1/32(you just need to be on the correct "turn" of the leadscrew"
Cut your thread, turn the lathe off, open half nuts, move the carriage back 7.5" close halfnuts, cut thread, repeat.

I cant check my maths on the lathe as its metric.

IF the above idea works, another idea you could make up a thread chaser dial with a 60 tooth* gear, wouldn't that let you pick up on the fly?

Granted this wont exactly be fast as you'll effectively be cutting a 7.5" long thread regardless of how much you want.(though you do get quick return unlike full reversal, if the chaser works you wouldn't even be turning the lathe off)

Stuart

p.s. Surely I'm missing something?

0.5 pitch = 2.5" reset
0.25 pitch= 1.25" reset

0.5 pitch = 20 tooth gear
0.25 pitch = 10 tooth gear

p.p.s I've got to be missing something!

No, your dead right. Just did some testing. Of course I have a 4tpi screw, but the distances stay the same, the number of turns is halved....I don't quite get that. I didn't even measure the distances though, I just counted the number of turns of the screw with the counter.

It may only work with a 127 tooth change gear, the math may be different for any other gears.

Pitch of thread to be cut x 127/pitch of leadscrew in metric (ie 3.175). The resulting number is the number of turns of the leadscrew between pick up points. I left the lathe running and just wound the saddle back to the start of the thread and waited for the right amount of numbers to go by, starting as soon as I dissengaged the half nuts. I'd does take a while even at 200rpm.

Cheers,
Ew

[Oldneweng]

I got the 127 tooth gear with my lathe. This is swapped for the 120 tooth gear for metric threading. Pretty standard on a lot of lathes.

My clutch/brake setup looks identical to yours except for one slight difference. You appear to have a lot more thickness of friction material on the brake disk. I have a piece of leather (I think) about 2.5mm thick. What is yours made of do you know?


What gears are needed Vernon/Dean? And what is the dp?

Vernon needs a 127 tooth gear.

Dp. Beyond my current ability to calculate. Have measured the following.

Outer diam 273.05. I used my 12 inch dial caliper for the first time for real!!:2tsup:

Root circle diam 264.414.

127 teeth.

Dean

[Ueee]

12 dp (OD in imp is 10.750", 127+2 gives you the OD of the gear, 129/10.75 is 12), i don't think i have a cutter.
Is that the only gear you need? Mine has 30/35/45/54. All the imp threads are cut with the 30 as the stud gear and the 45 as the gearbox gear. The metric are a combo of all 4, plus the 120/127t gears.

I don't even think i have a lump of steel/CI that big, Vernon if you can find me one (maybe a barbel weight, or weld up a ring and add a hub and spokes like Michael has done before) i'd be more than happy to cut you the gear. I'm sure i could find a 12dp No 2 cutter easy enough.

Cheers,
Ew

PS, the clutch is identical to the Nuttal branded clutches on my shaper and LeBlond lathe. Mine have some sort of black lining, bit like brake pad material, probably asbestos....

[Stustoys]

Hi Ewan,
Thanks, good to hear my maths still works now and then lol

I'd have to think more about "close approximation" change gears. Current thinking is, as long as 127*pitch is closer than say 1/4 the pitch of the leadscrew over the reset distance, you will still get repeat pick up.............not sure.

As your formula show, The pick up points are "Pitch of thread to be cut x 127" so the distance stays the same, just the number of turns changes for different leadscrew TPI

Great idea to use the thread chaser as is.

Its just occurred to me the time would be the same regardless of pitch, even though the reset distance is shorter. 127 turns = 127 turns. Just at finer pitches the leadscrew turns slower over a shorter distance.

Now......... is this any use on a metric leadscrew lathe?

Stuart

[Stustoys]

Hi Ewan,
Thanks, good to hear my maths still works now and then lol

I'd have to think more about "close approximation" change gears. Current thinking is, as long as 127*pitch is closer than say 1/4 the pitch of the leadscrew over the reset distance, you will still get repeat pick up.............not sure.

As your formula show, The pick up points are "Pitch of thread to be cut x 127" so the distance stays the same, just the number of turns changes for different leadscrew TPI

Great idea to use the thread chaser as is.

Its just occurred to me the time would be the same regardless of pitch, even though the reset distance is shorter. 127 turns = 127 turns. Just at finer pitches the leadscrew turns slower over a shorter distance.

Now......... is this any use on a metric leadscrew lathe?

Stuart

[Oldneweng]

12 dp (OD in imp is 10.750", 127+2 gives you the OD of the gear, 129/10.75 is 12), i don't think i have a cutter.
Is that the only gear you need? Mine has 30/35/45/54. All the imp threads are cut with the 30 as the stud gear and the 45 as the gearbox gear. The metric are a combo of all 4, plus the 120/127t gears.

I don't even think i have a lump of steel/CI that big, Vernon if you can find me one (maybe a barbel weight, or weld up a ring and add a hub and spokes like Michael has done before) i'd be more than happy to cut you the gear. I'm sure i could find a 12dp No 2 cutter easy enough.

Cheers,
Ew

PS, the clutch is identical to the Nuttal branded clutches on my shaper and LeBlond lathe. Mine have some sort of black lining, bit like brake pad material, probably asbestos....

You could have told me you wanted the figures in imperial and I wouldn't have converted them.:wink:

This is the Gear Train Ewan. 120 tooth at the bottom(Duh). The centre gear adjusts up and to the left to allow for the extra diam. That is all the external gears.

291844

Dean

[cba_melbourne]

Hi Ewan,
Thanks, good to hear my maths still works now and then lol

I'd have to think more about "close approximation" change gears. Current thinking is, as long as 127*pitch is closer than say 1/4 the pitch of the leadscrew over the reset distance, you will still get repeat pick up.............not sure.

As your formula show, The pick up points are "Pitch of thread to be cut x 127" so the distance stays the same, just the number of turns changes for different leadscrew TPI

Great idea to use the thread chaser as is.

Its just occurred to me the time would be the same regardless of pitch, even though the reset distance is shorter. 127 turns = 127 turns. Just at finer pitches the leadscrew turns slower over a shorter distance.

Now......... is this any use on a metric leadscrew lathe?

Stuart


Stuart, albeit I admire your math skills.... but.... is it not more convenient to leave these half nuts closed until the thread is finished, and turn this non-power-reversing spindle back manually the few turns between passes? Unless you are cutting a longish "allthread" section, it is probably faster turning the spindle back by hand, than it is to open the half nuts and then having to do your maths and rotate the spindle up to 127 turns to find the next sync point. Also consider that if you leave the half nuts closed, you avoid the considerable risk of syncronizing to a "false" (false = looks right but may be slightly off) pick-up point. Chris

[Vernonv]

12 dpThat's what I also worked it out as after Pipeclay gave me the formula via PM.

The other possible issue with getting gears made is the odd star shaped hub, which I would imagine would be a little difficult to replicate.

I'm going through now what gear combinations might be required to be able to create metric threads.

I'll get pics of the brake surfaces and the gear hubs when I get to the shed later.

[Vernonv]

OK, I have been back at the shed. Here are some pics of the brake setup. First one is the metal face, which I think is cast into the assembly cover. The second is of the friction material.

291905291906

The next set of pics shows the gears and their odd star shaped hub. Is that hard to replicate?
291907291908

So I've done some calculations on the gears required for metric threads from 0.5 up to 4.5. With the addition of 127, 30 and 35 gears I cover all thread pitches at an error of <0.00001% (a lot are actually spot on). In fact I can drop the 127 and still achieve an error of <1% on most except for 1.25 and 2.5 (any multiples of 1.25) which are <2%.

One thing I have also worked out is that none of the "new" gears would actually need to have the star shaped hub as they are all idler gears - so they could just have a round hole to fit the idler post (see pic above).

[Michael G]

Yes, that star shaped hub will cause some angst. I think I'd be drilling 8 holes on their PCD (bit of layout work required) then boring the main hole and finally finishing up the star with a file or two.
That arrangement is only for keying the 120 to 127 gears together. An alternative may be to make up a new 120t gear as well and use a normal key and sleeve. The fist option is probably slightly less work but will require more hand work than the second.

Michael

[Vernonv]

That arrangement is only for keying the 120 to 127 gears together.Certainly the case for the idler gears, however that star arrangement is also used on the spindle and QC input shafts.

I figured it would be a simple task just to blind drill a hole into each of the gear hubs, all at the same distance from the centre and use a pin to key the shafts (i.e. have the holes facing each other).

[Oldneweng]

The star shaped shaft is part of the driven shaft on my lathe. It does not come off. Could a 127 tooth gear be made with just a bore to go over the OD of the shaft and then pin the two gears together.

Dean

ps How does the star hub bit lock on to the driven shaft Vernon.

pps I just realised that you are showing the 60 tooth gear.

Is your 120 tooth shaft solid as per mine then?

[PDW]

I've got a set of 12DP 14.5 PA cutters. I'll lend the appropriate one out to someone if they want to cut a gear or 2. ATM I don't have my dividing head set up and I'd have to do direct indexing off of a 127T gear anyway, which is a PITA. I'm also short on time and about to go travelling (again).

Hopefully this time less than 3 tonnes of machine tools will accompany me home.

Anyway the offer of the cutter is there...

PDW

[Oldneweng]

Okay, so my idler gear is 72 tooth not 60 which anyone could see if they looked close at my gear train photo. The gear has 72 cast into one spoke.:C

My drive gear is 30 tooth.


With the addition of 127, 30 and 35 gears I cover all thread pitches at an error of <0.00001% (a lot are actually spot on).

Do you actually need a additional 30 tooth gear Vernon or was that a typo?

Dean

[Stustoys]

Stuart, albeit I admire your math skills.... but.... is it not more convenient
I say there is a fair difference between "there is absolutely NO way to disengage the half nuts without ruining the thread being cut" and "more convenient".

There is no maths to do while threading*...... do it once and its done.
You're using the thread chaser so an error is no more likely than "normal" threading. If you like, the pick up can be done with the lathe off.

Whats more convenient depends on a number of things, Length of thread and number off would be the big two but the lathe comes into it, as does laziness:D. I only recall cutting one thread by hand, 6BA I think it was and only about 8mm long, winding my lathe over by hand isnt something I enjoyed.. and my lathe is tiny compared to some here. A handle on the back of the spindle wouldnt be great as you would have to move around the lathe to get to it(though again it would depend). I doubt turning a Nutalle backwards would be anymore enjoyable. YMMV
Being able to open the half nuts may allow you to run the lathe faster so the time between pick ups would be less. Though even at 200rpm its 30 odd seconds - the time to get ready for the next cut - the thread length(in time) - the possible time saving of cutting the thread at a higher speed.

Even if the lathe has reverse being able to open the halfnuts may reduce the pucker factor of threading to a shoulder.


Stuart

*unless you call counting turns of the thread chaser maths. With a custom gear that could be removed.

[Vernonv]

How does the star hub bit lock on to the driven shaft Vernon.Not sure. I had a look but it was not obvious.


I just realised that you are showing the 60 tooth gear.72 actually


Is your 120 tooth shaft solid as per mine then?If you mean the driven shaft, yes it is solid.

[Vernonv]

Do you actually need a additional 30 tooth gear Vernon or was that a typo?No, the extra 30 is for 1.5mm pitch and multiples.

[cba_melbourne]

> I say there is a fair difference between "there is absolutely NO way to disengage the half nuts without ruining the thread being cut" and "more convenient".

Hmm, I would say in this case it is sooo much more convenient, that trying to do anything else is pure masochism :))

> Being able to open the half nuts may allow you to run the lathe faster so the time between pick ups would be less. Though even at 200rpm its 30 odd seconds - the time to get ready for the next cut - the thread length(in time) - the possible time saving of cutting the thread at a higher speed.

If worried about saving time, I think the OP would have added a spindle reverse a long time ago.
I concede you may be right if the OP did ask about cutting a large batch of threads of same pitch. But if he is only after a one off or two, I bet in his particular situation it is at least 20 times faster to do it by leaving the half nuts closed and leave the maths alone.


> Even if the lathe has reverse being able to open the halfnuts may reduce the pucker factor of threading to a shoulder.

I can agree with this, but would take it one step further. Why only reduce the "threading to a shoulder" pucker factor, when it can easily be removed altogether?
Just run the spindle in reverse and use the threading tool upside down. That feature alone would make adding a spindle reverse worthwhile.


Chris

[Vernonv]

Just run the spindle in reverse and use the threading tool upside down. That feature alone would make adding a spindle reverse worthwhile. My lathe has a threaded spindle nose, so that may not work very well (in my situation anyway).

[Oldneweng]

No, the extra 30 is for 1.5mm pitch and multiples.

I should explain myself better. I was running in and out this morning. Out to the shed and back in again to type just to find I had been beaten to the punchline.

My gear train is 30/72/120 or 30/72/127. Sorry about the "60" but I did correct that in post #49:(.

I could not work out the reason for an additional 30 tooth gear unless you had a different gear altogether. As far as I can tell (this means not much) there is only one position at a time for a 30 tooth gear. This is why I asked.



With the addition of 127, 30 and 35 gears I cover all thread pitches at an error of <0.00001% (a lot are actually spot on).




In addition to which gears other than the idler and 120?


As per your answers to my questions in post #51 just pretend you did not see my questions. I managed to work out what was going on eventually.

This is all academic to me as my lathe is moving to a new home eventually but it is still something that I want to learn. I realised that today trying to get my head around this issue.

I had a quick look at the CY lathe manual today to see what sort of arrangement it has for threading and quickly decided that could wait until it has been installed. I am sure it is easier than it looks at first glance but I want to wait until I am able to put the info into practice.

Dean

[Vernonv]

I could not work out the reason for an additional 30 tooth gear unless you had a different gear altogether. As far as I can tell (this means not much) there is only one position at a time for a 30 tooth gear. This is why I asked. The extra 30 would be used in a compound situation - for example for a 1.5mm pitch you would have the driving 30 driving the 127 and 30 in compound configuration on the idler, with the 30 driving the 120 in the driven position.



In addition to which gears other than the idler and 120?The way I've worked it out, the (original) 30 will almost always be in the driver position, whereas the 120 or the 72 must be in the driven position. The additional gears (127,30, 35) will only ever be used in the idler/compound position, as they will not have the correct bore configuration (they will be smooth bore, rather than the star type).

I hope that all makes sense.

[Oldneweng]

The extra 30 would be used in a compound situation - for example for a 1.5mm pitch you would have the driving 30 driving the 127 and 30 in compound configuration on the idler, with the 30 driving the 120 in the driven position.


The way I've worked it out, the (original) 30 will almost always be in the driver position, whereas the 120 or the 72 must be in the driven position. The additional gears (127,30, 35) will only ever be used in the idler/compound position, as they will not have the correct bore configuration (they will be smooth bore, rather than the star type).

I hope that all makes sense.

Thanks for clearing that up Vernon. Perfect sense now that you have explained it. I have doubts that my lathe is good enough now to use for threading to any accuracy but it is all necessary knowledge for future use. I worked out tables for the gear combinations I have plus a 35 tooth just for fun but did not go any further. It became obvious that the 1.25 combinations were an issue.

Dean

[Stustoys]

[QUOTE=cba_melbourne;1713160 I bet in his particular situation it is at least 20 times faster to do it by leaving the half nuts closed and leave the maths alone. [/QUOTE]

I had doubts about the 20 times number, maybe 2, maybe less, it depends how easy your lathe happens to be to turn over by hand. I was going to try it out and see so thought I would set my metric lathe up to cut an imp thread using the same idea. Now assuming my maths is correct I believe this time we can agree its unlikely to be much use. The pick up points are 15inchs apart(dont really have my head around that yet, Ewan come pull my maths into something I can understand lol, its 127 again but its in a different column:no:)

Back to cleaning

Stuart

[Ueee]

My head hurts.......let me think on it for a while. I should be doing bookwork right now.

[Ueee]

Whilst this does not help the imp on a metric lathe directly, it occurred to me that the time taken for the leadscrew to spin its number of times over before half nut re-engagement is the same no matter what thread (i haven't looked at odd pitches yet though) you cut. The carriage movement may change, as may the number of turns but there is one constant-the number of times the spindle revolves from one engagement point to the next. Without doing any maths i bet the number is 127 times.

Stuart, What pitch leadscrew did you work that excel sheet out on?

Chris, unless its a long thread there is no way i'd do this over reversing. For the last 8 odd years reversing is the only way i have cut threads. It would be way slower on short threads. Thanks for the tip of cutting from a shoulder in reverse, i'll remember that one!

Ew

[Stustoys]

Hi Ewan,
I'd rather your head hurts than mine ;)
No. cutting metric on imp screw, the pick up distance gets shorter so the time remains the same. But cutting imp on metric screw the pick up distance seems to stay the same and the time increases as pitches get finer. 300 pitch for 20tpi, 600 pitches for 40tpi!

Sorry I should have labeled

The first is 1.5mm pitch thread on an 8tpi leadscrew
then 0.5mm on 8 tpi
0.25mm on 8tpi
40 tpi on 3mm lead screw
20 tpi on 3mm

Stuart

[cba_melbourne]

Thanks for the tip of cutting from a shoulder in reverse, i'll remember that one!

Ew


For outside threads it is easy, you can just reverse the cutting tool upside down in the standard toolholder and reverse the spindle. One would intuitively think this cuts a left hand thread; but you get indeed a normal right hand thread with the tool moving towards the tailstock. If you have a rear toolpost, you can of course also clamp the tool in there, not upside down, and run spindle in reverse. You need to make a relief at the shoulder though, as you have to somehow advance the tool before the cut starts. If you had a threaded spindle nose, the chuck could spin off under a heavy cut. I do not think the forces when cutting smaller threads say M12 and smaller would undo the chuck. But larger threads maybe yes. There are two methods, one cheap and nasty is to add a grub screw to the back plate, that locks it to the spindle thread. But the grub screw may over time damage the precious spindle nose, especially so if the spindle is not hardened. The better way is to make a new backplate with a locking collar, but its much more work - and if the type of spindle relies entierly on the threads to center the chuck, not on a tight fitting register, the locking collar may actually affect chuck runout.

But the greater problem when cutting to a shoulder are actually the internal threads. The angst factor is greater, because not only is there a shoulder, you additionally cannot see the shoulder. Just flipping the tool upside down and run the spindle in reverse as for external threads will not work here. You really need an internal threading tool that is a mirror image of the normal tool, a left handed threading tool. Chris

[Stustoys]

Here you go Ewan. Have been showing my working.
I think the maths to get the leadscrew turn is
1/(3/(25.4/14))

[Ueee]

I see where i stuffed up....my head is now unexploded.....

So to work out the turns of the spindle you need 127/(pitch in metric/pitch of leadscrew) divide that by the tpi to get the saddle reset distance. For 20 tpi its 127 pitches of the leadscrew or 15 inches. Actually for a 3mm screw it seems to be always 15"/127 pitches....works for odds....for half threads it needs to be 254/, so the reset distance is double.
For different leadscrew pitches it is still 127 (or 254) pitches, so the distance for reset gets bigger. 6mm pitch (whats a standardish metric leadscrew pitch? The Mars is 3.5mm) and you need 30" to reset......

The number of spindle turns for metric on an imp lathe is 127 too. So although it seemed like ages it really only took roughly 40 sec at 200rpm, minus some for moving the saddle back.

Chris, All the internal threading i have done with a HSS bit in a boring bar, so reversing it to cut in reverse would not be a problem. This only works if threading into a shoulder and not into a completely blind hole as the end of the bar would get in the way.

Cheers,
Ew

[Michael G]

But the greater problem when cutting to a shoulder are actually the internal threads. The angst factor is greater, because not only is there a shoulder, you additionally cannot see the shoulder.

I cut them 'normally' but put a piece of masking tape on the bar holding/ forming the cutter. I then know that as soon as there is no metal visible between the tape and the end of the job I need to flick the cutter in and disengage feed. (Having a lathe with the single tooth clutch I cut threads with the half nuts always engaged anyway). It takes some getting used to but the tape is a surprisingly good depth indicator.

Michael

[Stustoys]

Hi Ewan,
Thanks for that.
So cutting metric pitches on an imp leadscrew. Pick ups are 127 turns of spindle apart so pick up distance changes with pitch but the time between pick ups stays the same.(38 seconds at 200rpm)

Cutting imp threads on a metric leadscrew. Pick ups are 127 turns of the leadscrew apart so pick distance stays the same(15") and the time between picks changes. (about 1 minute for 14 tpi at 200rpm which depending, one might be able to live with. 3 minutes for 40tpi not so much)

3mm is standard for us little guys. certainly 6mm on larger lathes but I have no idea what others are used (I would never have guessed 3.5mm)


Stuart