Threading

  1. Agroshield Member

    Messages:
    1,389
    Another couple of ways that I have read of:

    Set compound parallel to lathe axis and move forward one pitch. Means you can only screwcut 'straight in'.

    If you are using change gears, mark two or three evenly-spaced teeth on the mandrel gear and one mark on the meshing gear. Remove, index to next mark. Replace.

    Monarch made a 'multiple index face plate' specially for the job. Have a look at page 14 here:

    https://www.sterlingmachinery.com/m...itive-precision-toolmakers-lathe-brochure.pdf

    Edited to add:

    Look at post #15 here if you have tumbler gears:

    https://www.practicalmachinist.com/vb/monarch-lathes/ok-more-unknown-tooling-monarch-10ee-371035/

    Here, it says if you are turning between centres, just index the dog on the chuck jaw. If you decide to cut a 2-start and use a 3-jaw chuck, the process is a little challenging:

    https://www.practicalmachinist.com/vb/general-archive/threading-question-158660/
     
  2. Dr.Al

    Dr.Al Forum Supporter

    Messages:
    784
    Location:
    Gloucestershire, UK
    When I turn between centres I tend to use a soft centre in the chuck (turned from some hex bar I had lying around) and turn the point just before use so it's super accurate. I then have a drive plate that is mounted to the hex bar with tapped holes for a bolt that acts as a driver:

    [​IMG]

    You can see two tapped holes for doing two-start threads, but since taking that photo, I've needed to cut a few multi-start threads so have just drilled/tapped some more holes in the face to give me whatever number of starts is required. It's now got quite a lot of (labelled) holes in it! Just have to get it centred nicely on the mill (with a DTI) before drilling & tapping the holes (with the DRO's PCD function to make it easy).
     
    Brad93 likes this.
  3. Dr.Al

    Dr.Al Forum Supporter

    Messages:
    784
    Location:
    Gloucestershire, UK
    Image doesn't seem to be working - try this:

    upload_2020-6-12_8-27-44.png
     
  4. zx9

    zx9 Member

    Messages:
    4,334
    Location:
    South East London
    Is there a reason why you are not just using a centre in the lathe's spindle with the catch plate screwed on to the nose?
    Using the chuck you are just moving everything along the bed, it is not like you have a gap to avoid, or am I missing something.
     
    Seadog likes this.
  5. Agroshield Member

    Messages:
    1,389
    What struck me about that and is almost an unseen part of its cleverness is using the hex bar. This both avoids having to have a chuck whose jaws are a factor of the number of starts and means your driver plate only needs one driver location (stud for straight dog or slot for bent dog). You just clamp it on a different flat for each thread start.

    E.g. if using a 3-jaw and wanting a 4-start, use a bit of square bar: one end turned round to grip in chuck, the other turned insitu to a point. If you want a 5-start, mill up a pentagon and away you go.
     
  6. Pete.

    Pete. Member

    Messages:
    9,817
    Location:
    Kent, UK
    You can do that regularly but his compound doesn't have the travel to move 2 pitches, so he couldn't use that trick. The gear tooth method would work though so long as his stud gear was divisible by three.

    This would be ok for a rough thread but your 3 leads would only be as concentric as your 3-jaw chuck could hold the hex. In fact, any runout when you turn the point would be almost doubled as you move the hex around.
     
  7. Seadog

    Seadog Forum Supporter

    Messages:
    6,976
    Location:
    NE London - UK
    You'd need to re-cut the taper every time which would then introduce an error longitudinally.
     
    Brad93 and Dr.Al like this.
  8. Dr.Al

    Dr.Al Forum Supporter

    Messages:
    784
    Location:
    Gloucestershire, UK
    A few reasons:
    1. My lathe is plenty long enough that I have never needed the extra length that I'd get by removing the chuck
    2. A centre in the lathe's spindle (which would probably be hardened) wouldn't be as accurate (especially as it would have to be in a home-made MT4.5 to MT3 adaptor) than one turned in place immediately before use.
    3. A catch plate fitted to the nose would require me to buy a dedicated D1-3 back plate to make into a catch plate and my way is cheaper as it used scrap bin stock.
    It's a nice idea, but it would mean re-cutting the taper to get the accuracy (any inaccuracy in the chuck would cause a problem otherwise). It would also introduce a longitudinal offset if I re-cut the taper, although that's wouldn't happen if I didn't re-cut it as I tend to push the drive plate up against the jaws.

    The main reason I used hex bar was that I had a bit in a weird (for which read "imperial") size that came out of the skip and since it's an imperial hex size I wasn't likely to use it for anything else. It does have the advantage that (unlike a bit of round bar) there's no chance of it rotating in the chuck and hence losing thread registration if I haven't tightened the chuck enough.
     
    Shox Dr likes this.
  9. Hopefuldave Intergalactic pot-mender

    Messages:
    1,805
    Location:
    The Shed of Danger, surrey, England
    If the stud gear on the spindle is divisible by the number of starts, cut one of the starts/threads, then chalk a line across where each pair of gears mate - then disengage 'em, rotate the stud (spindle) gear by turning the chuck though "teeth divided by starts", re-engage with the chalk marks lined up, cut the next start, rinse and repeat - a bit old-school, but works!
    ( I have done 3-start like this, with the gears on the lathe I can do 2, 3, 4, 5, 6 - and 9, 12, 15, 18 - but I don't want to!)

    Dave H. (the other one)
     
  10. Hopefuldave Intergalactic pot-mender

    Messages:
    1,805
    Location:
    The Shed of Danger, surrey, England
    I got around that, a spare* camlock pin and a fat sleeve threaded inside to take the pin's thread, lock the pin in the spindle nose then screw the sleeve down, instant drive pin...

    Dave H. (the other one)

    * D1 backplate came with six, three's plenty for the little (5") chuck
     
  11. Dr.Al

    Dr.Al Forum Supporter

    Messages:
    784
    Location:
    Gloucestershire, UK
    I was thinking about this again and I think your idea could work, just in a different way. The circular drive plate that I showed in the photo is attached to the hex bar with three cap screws that go all the way from the outside diameter of the drive plate and bear on the hex bar. The inner diameter of the drive plate is the same diameter as the across-corners dimension of the hex bar so it's a close sliding fit on the bar. If I were to undo the three cap screws a few turns, I could index the drive plate round such that the cap screws bear on a different face of the hexagon without moving the hex bar itself and hence keeping everything square. That would be really easy to do and would be a simple way to cut 2, 3 or 6 start threads when turning between centres.
     
    Pete. likes this.
  12. Agroshield Member

    Messages:
    1,389
    Yes, and that is the genius of the method you show. You might need a circular register on the chuck end of the bar that helps locate the drive plate concentrically.

    I am not sure where the idea of longitudinal error comes in. You would surely not remachine the centre between each start of the thread. Yes, over the years it would become shorter, but that is not a problem if it starts out long enough in the beginning. When you index the drive plate, the stock just rotates on the centre that has been machined specifically for the thread you are now cutting; it does not move fore or aft.

    To Pete. I also do not think the hex necessarily needs to be concentric. You are not removing it from the chuck at any time. You are clamping the drive plate to a different flat on it for each start. For a four start thread, a rectangular bar could be used. As long as it has four sides at 90 degrees to each other, it would work. All that would change is the amount you wind the clamping screw in to lock the drive plate onto the flat.

    The hexagon (or octagon or decagon) bit could even be made as a nut that threads onto a universal central carrier that is made ten years long to begin with. You would just have to devise a sure way that the nut could not unscrew itself at the wrong moment.
     
  13. Pete.

    Pete. Member

    Messages:
    9,817
    Location:
    Kent, UK
    I do beg your pardon I mis-read your post to mean that you could turn the hex in the chuck to index the three starts. Turning the drive plate on the hex would work, but you'd have to make sure that the screw was right in the middle of each flat to minimise indexing error. Just a bit of care when setting up would do it.

    You could even use a round bar and round hole in the drive plate, and drill the bar for indexing on a dividing head.
     
  14. slim_boy_fat

    slim_boy_fat Forum Supporter

    I do it the scaredy-cat's way! :ashamed:

    Workpiece in the chuck, drill a start with a centre drill in an R32 collet in the tailstock and advance into the workpiece while turning the workpiece by hand. Then work upwards until I get to the appropriate drill for the tap size. That way I can back it off as said, faster than faffing about with gears etc. :ashamed:

    I learned about backing it off when I tried for the first ever time to tap something - a piece of ali flat bar with to lube or back off. :doh:
     
  15. Kram

    Kram Member

    Messages:
    1,786
    Location:
    Sussex


    I found this interesting, needs to be a coarse multi start for it to work? Im not sure why he used router to mill the threads.
     
    zx9, cocker and Shox Dr like this.
  16. Agroshield Member

    Messages:
    1,389
    For something similar, have a look at self-reversing screws. See also Yankee screwdrivers and the device for spooling cable neatly onto a drum.

    I have a vague idea the idea dates back a long time, to when mechanical cloth weaving started, for loading the thread onto the shuttles.

    It would be hard to get sufficient surface speed with that using a single point tool for a good finish. Once you started the other hand thread, it would be a very interrupted cut.
     
  17. Seadog

    Seadog Forum Supporter

    Messages:
    6,976
    Location:
    NE London - UK

    Probably the only high speed lump he could attach to his toolpost.
     
    Hopefuldave likes this.
  18. ranchero Member

    Messages:
    132
    Location:
    southampton
    used to be our test piece when i was a apprentice [ 65 now] wish i still had mine as i could not do it now
    just for looks we used 3 different metals and a different knural on each nut
     
    slim_boy_fat likes this.
Advertisements