Announcement

Collapse
No announcement yet.

Bison chuck internal thread not fully cut?

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • old mart
    replied
    Try putting a piece of steel bar in your four jaw chuck, about 2 to 3 inch diameter, projecting about 3 inches and turning it true to the spindle. Clamp the Bison onto the bar and you can check any runout present.
    There may be a part number on the chuck to confirm that it was or was not intended to fit your lathe.
    Last edited by old mart; 09-02-2017, 04:09 PM.

    Leave a comment:


  • danlb
    replied
    For a new quality chuck .004 is a lot for a 3 jaw. Is it wobble or is it off center? Wobble is .004 at the jaws and .010 when measured 3 inches out. Off center is .004 at the chuck and close to .004 at 3 inches out.

    Wobble may mean that the end of the "registration collar" is not perpendicular to the axis of the chuck. It could also be due to damaged jaws (bellmouth) and several other causes.

    Off center can mean the jaws are not closing uniformly (could be wear or could be swarf in the scroll gear or even swarf in the gripping part of the jaws) or that the registration bore is not centered on the axis, thus pulling the whole chuck off center.


    Dan
    Last edited by danlb; 09-02-2017, 03:48 PM.

    Leave a comment:


  • old mart
    replied
    Try checking the wobble after tightening each pinion in turn, you may be able to reduce the wobble. Checking with different diameter stock in the jaws would find out whether the wobble is consistent.
    Check the trueness of the face and rear of the chuck, and the outside diameter.
    Last edited by old mart; 09-02-2017, 03:41 PM.

    Leave a comment:


  • danlb
    replied
    Ok, a bunch of people say the register is not used. I don't agree. There would not be a boss on virtually every screw type spindle nose if it was purely decorative. When I look at the spindle on mine, the boss definitely makes good contact as the chuck's spigot/flange contacts the face of the spindle.

    Have any of you had reason to cut away the boss to see what difference it makes? I'd love to hear of empirical proof that it makes no difference to the centering of the lathe.

    As a test of the need for a boss; I marked a line on my chuck body near the jaws for reference and used a dial indicator to find that it had about .001 TIR on the outside of the chuck. Repeated several times with the same results.

    Then I unscrewed the chuck a bit and inserted a matched pair of 1/2 inch parallels as shims between the spigot and the face of the spindle so the boss was not touching the chuck. I tightened it down so both parallels were trapped and tested again at the same spot. Now I have almost .017 TIR. Repeated several times with the similar results (.012 to .019 TIR) . The high spot moves, BTW. The boss makes a very significant difference on mine.



    Dan

    Leave a comment:


  • Graydon
    replied
    HI Guys,

    Thank you all very much for the many fine points and suggestions. Also, to correct what I said in the original post - the bottom of the thread gauge in the internal threads was not really visible so I can't really say there was daylight showing at the root.

    Based on the advise from all of you I first shaved a few thou off the sharp V threads on the Logan lathe spindle. That allowed me to thread the chuck onto the spindle until it contacted the vertical face of the register. Then I measured the outside diameter of that "register" at 2.286".

    Then I mounted up my 4-jaw chuck and then chucked the Bison in my 4-jaw chuck. I indicated it as closely as I could (less than 0.001" runout) and then turned out the inner diameter of the Bison to clear the registration collar. It ended up around 2.287". The Bison now fits and gives a solid "chunk" when the backside of the Bison hits the registration wall. I put a piece of ground shaft into the Bison and I get roughly 0.004" wobble. Is this normal and expected or should I remove it and look for crap getting in the way on the registration collar?

    Thank you all once again.

    Sincerely,
    Graydon Stuckey

    Leave a comment:


  • old mart
    replied
    We have 2 faceplates, 4 three jaw chucks and 2 four jaw chucks for the Smart & Brown at the museum, all use backplates. About half of the backplates were inherited or obtained readymade, and the others were produced from blanks by me.
    Some of mine are tight enough to require a little lubrication when screwing on and the others vary in fit. The sloppiest one has about 0.012" clearance right up to final tightening, in the threads and register. As the faceplates are finally machined on the spindle before assembly, they all run equally well.

    Leave a comment:


  • MattiJ
    replied
    Originally posted by J Tiers View Post
    Never heard of threads that would have light at BOTH crest and root with a full V thread gage. BUT there IS a way you could see that.

    If the thread gage has one side trimmed for the root flat, AND the chuck has full V roots but flats on the ID crests, then you would see what is described.

    The gage only needs the flats on the part that reaches the root, for cases where the thread root is flattened. The root does not HAVE to be flattened, it is "allowed". The crest HAS TO be flattened, to fit cases where the root is flat. Both on gages and threads.
    Yeah, that's why I was asking since OP stated that "The thread on the lathe spindle matches my thread gauge perfectly. I checked the thread on the chuck and I can see daylight at the tip and root of the thread gauge."

    Withworth thread, too large pitch diameter?

    Leave a comment:


  • CalM
    replied
    Originally posted by J Tiers View Post
    Yep.

    If the round part behind the threads were intended to hold it concentric, it would be a taper. With a parallel fit, it would have to be a tight "wringing" fit, and would quickly wear out of tolerance. Even one thou of slop all around would bring the chuck well out of being concentric.

    Also, the threads DO have a centering action, so the two, threads vs the so-called-by-many "register" would fight one another unless they were perfect. But if the threads were perfect, the "register" would be un-needed.

    Numerous fine machines have a recess there, no "register" at all. Sherline is one.

    The whole "register" thing is just something for the armchair machinists to argue about.
    Entirely CORRECT!

    Think of the thread as multiple taper fits not so very dissimilar from a collet. What the thread has going for it is that it is self closing, and since there are several in series, there is a very strong averaging taking place. One defect doesn't upset everything.

    The whole "register" thing is just something for armchair machine whores to argue about.

    Leave a comment:


  • J Tiers
    replied
    Yep.

    If the round part behind the threads were intended to hold it concentric, it would be a taper. With a parallel fit, it would have to be a tight "wringing" fit, and would quickly wear out of tolerance. Even one thou of slop all around would bring the chuck well out of being concentric.

    Also, the threads DO have a centering action, so the two, threads vs the so-called-by-many "register" would fight one another unless they were perfect. But if the threads were perfect, the "register" would be un-needed.

    Numerous fine machines have a recess there, no "register" at all. Sherline is one.

    The whole "register" thing is just something for the armchair machinists to argue about.

    Leave a comment:


  • firbikrhd1
    replied
    I agree with lakeside and would like to mention that in addition, the portion of the threaded area that butts against the flat shoulder on the spindle that is perpendicular to the axis of the lathe has a great deal to do with whether the chuck will run true. In other words when the chuck is screwed on completely the back of the threaded portion must meet (butt) the spindle equally around the entire circumference of the shoulder. A chip between or a ding on either surface will make the chuck run out of true.

    Leave a comment:


  • lakeside53
    replied
    The recesses are almost always oversize and on some chucks almost non-existent. The screw action when tight pulls the chuck into axial alignment. Sure.. it's "lose" until tightened, but when tentioned to the vertical register the thread form acts as a cylinder.
    Last edited by lakeside53; 09-02-2017, 12:23 AM.

    Leave a comment:


  • danlb
    replied
    Originally posted by lakeside53 View Post
    Not true.


    The spindle is aligned by the threads and the chuck pulled against the the vertical face of the register. The horizontal area (shoulder) is not a register. The thread pitch is concentric to the center line (of course) and that is the register.

    The only time the shoulder matters is if you have a clamping device that locks the chuck to that like EMCO. In that case you can pull the chuck out of alignment if the shoulder is not a close fit.
    It has been my experience that the flank of a screw thread is almost never a valid register. The face of the spindle register will guarantee that the axis of the chuck is parallel to the axis of the spindle. The clearance needed to keep a thread from binding and galling will cause wobble and let the two pieces shift from side to side. The chuck recess that matches the register guarantees that the axis is concentric.

    To prove it to yourself, take the best screw that you have and screw it onto it's matching screw without tightening it. If you use a pair of vice grips (to magnify the movement) you will see that even a high class nut will wobble on the screw.

    Dan
    Last edited by danlb; 09-02-2017, 11:25 AM.

    Leave a comment:


  • thaiguzzi
    replied
    Could be you have an 8 TPI American 60 degree thread and the chuck has a Brit 8 TPI 55 degree thread. Quite common in the UK to find a 9" Southbend 8 TPI x 1.5" backplate does not want to screw on easily onto a Boxford 8 TPI x 1.5" spindle. Difference being the 5 degrees.

    Leave a comment:


  • lakeside53
    replied
    Correct!

    Leave a comment:


  • Graydon
    replied
    Originally posted by lakeside53 View Post
    Not true.


    The spindle is aligned by the threads and the chuck pulled against the the vertical face of the register. The horizontal area (shoulder) is not a register. The thread pitch is concentric to the center line (of course) and that is the register.

    The only time the shoulder matters is if you have a clamping device that locks the chuck to that like EMCO. In that case you can pull the chuck out of alignment if the shoulder is not a close fit.
    Lakeside,

    OK, then in that case, the outer diameter of the shoulder doesn't really matter that much. I can machine the chuck with a few thou clearance to that shoulder and it should run true because the flanks of the threads are maintaining the concentricity of the chuck with the spindle.

    Is that correct?

    Thanks,
    Graydon

    Leave a comment:

Working...
X