Announcement

Collapse
No announcement yet.

D1-4 spindle, should chuck fit flush ?

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

  • D1-4 spindle, should chuck fit flush ?

    I notice when I fit my chuck there is a slight gap between the back of the chuck and the face of the spindle. The front part of the spindle has a slight taper which I suppose is to provide a taper type lock and then using the cam locks, pull the chuck in tight over the taper. Is this correct ? Or should the chuck fit flush on the spindle's face ?

    I'm still trying to solve the chuck runout problem, I think the chuck is not fitting correctly on the spindle mount.


  • #2
    It should fit on the taper firm and take a tap with a rubber mallet to remove, but should pull up against the spindle.

    Dave
    EDIT
    Just to add that I picked up a spare 3 jaw a couple of years ago and it was a tight fit and wouldn't pull up, so I mounted the 3 jaw in a 4 jaw backwards and machined the taper out a little. Doing it this way I could turn the whole lot around to try it on the spindle then machine a little more as needed until I got a good fit.
    Last edited by Davo J; 12-12-2010, 07:00 AM.

    Comment


    • #3
      Do you have (or have access to) something else on a D1-4 mounting? If so, try fitting it - as davo says, the flat faces should pull up and make contact. This will give a clue as to whether it's the spindle or backplate that are not within tolerance.

      Having a gap between the faces will play havoc with the chuck's alignment, and will be seen by taking runout readings close to the chuck, and then further towards the tailstock; the further from the chuck, the greater the runout.

      Ian
      All of the gear, no idea...

      Comment


      • #4
        [QUOTE=TR]

        I'm still trying to solve the chuck runout problem, I think the chuck is not fitting correctly on the spindle mount.

        [QUOTE]

        sounds like you've found your problem....now your problem is figuring out if it is the lathe or chuck that is incorrect. Lets hope its the chuck!

        The taper locates things, its not really a lock, but it as to be just so such that the chuck back is in full contact with it and the face of the spindle. Right now I'd bet one of the pins pulling it into contact making it off kilter like swash plate. If you need further proof that this is your problem indicate your test bar and see how wildly the readings change as you move away from the chuck

        Its a difficult thing to measure, where a cone meets a plane, and tricky to machine. I'd start by checking everything else you've got that fits. If nothing fits well, probably the lathe, if everything does, probably the chuck. Check fit with feeler gauges and with with an indicator - see if you can move the chuck side to side (checks if its engaged in on the taper). Some blue is also a good way to check.
        in Toronto Ontario - where are you?

        Comment


        • #5
          Since I've just got a machine with D1-3 camlock spindle, this is something I've been thinking about.

          I tend to agree that to leave everything to the taper, ie having a significant, say > 5 thou, gap between the faces, would allow uneven tightening of the cams to leave the chuck pointing off-centre.

          But where do you stop ? If the faces butt up properly, and the taper has a gap, you've got no centring.

          So ideally you'd like the clamping pressure shared between the taper, to get centring, and the faces, to get direction.

          Right. Well, that's a pretty fine tolerance if you ask me. So you've got to decide which side to err on, and I'd prefer to err on leaving a gap.

          Let's look at it another way. Suppose you leave a gap, and you tighten the cams harder. The taper clamps up harder and the gap gets smaller. So everything gets better.

          Now suppose there's no gap, and the taper is loose. Tighten it harder, and the taper will stay just as loose as before. All you'll do is make the lack of centring more definite.

          So, I'd rather have a very small gap, that tended to disappear as I tighten the cams a little too tight.
          Richard - SW London, UK, EU.

          Comment


          • #6
            If it only needs a small amount use a piece of emery while it's mounted backward like I said above instead of machining. Somewhere on this forum someone posted how much off the taper would move it closer to the spindle face (X) amount. I know it's not much and when I machined it, I over shot and then had to face off a bit to get it back again. I then finished it off with emery.
            I wouldn't touch the spindle and you will find it's hardened. All my chucks/back plates get pulled up with the cams and then it takes a swift blow from behind with a rubber hammer to unseat it off the taper and I have no repeatability problems.

            Dave

            Comment


            • #7
              The taper at the end nearest the spindle face should measure 2.50050 +0.0005. The specs are very clear on this. Here is a link to what things should be mearsuring out to....
              http://shopswarf.orconhosting.net.nz/chuckmt.html

              I have been making D1-4 Rotary table adapters for the D1-4 and I have made everyone to these measurements....all work perfectly.

              If it is the spindle I would go back to the seller and have them replace your lathe....
              Bob
              Pics of shop and some projects
              http://s2.photobucket.com/albums/y39...achine%20Shop/

              Comment


              • #8
                The D1 and A1 spindle nose series standards provides a little "draw" where the female taper is a trifle smaller than the male taper. The gage line dimensions of the tapers specifed in the standard show this interferance but I can't recall exactly what it is.

                If the chuck is mounted and held engaged but not clamped you should be able to get a 0.001" or 0.002" feeler in the gap. When the camlocks are drawn up to be fully tensioned, the taper engages, elastically deforms a trifle in acommodation, and the back face mates solidly. It's a very accurate, repeatable system if kept scrupulouly clean during tooling changes.
                Last edited by Forrest Addy; 12-12-2010, 10:45 AM.

                Comment


                • #9
                  Originally posted by deere_x475guy
                  [FONT=Arial][SIZE=3]The taper at the end nearest the spindle face should measure 2.50050 +0.0005. The specs are very clear on this. Here is a link to what things should be mearsuring out to...
                  clear on paper, difficult to measure or to do. Most spindles will have a relief there, or should, so the intersection of cone and face is a theoretical place not actual. Even if it were actual, any radius on the tool or grinding wheel that cut it will wreck the measurement....and what measurement tool do you get in there that is reliable to 1/2 a thou?

                  When I've done it I've used a female i trust (not many of those lol), turned the taper first, then start kissing off thou's on face using feeler gauges to know how much to take off. I didn't know until Forrest's post that it was supposed to be a thou but that's probably where it ended up.
                  in Toronto Ontario - where are you?

                  Comment


                  • #10
                    McGyver

                    Gage lines have to be measured on the male taper. You can't actually measure the gage line axial location of a flanged gage. You have to calculate the offset and tangency of pins from the gage line (intersection diameter) and measure over the pins. It's a real worm rassle to set up and measure ensuring the pins are shouldered and parallel to each other but it can be done. As you say, tapers are best gaged and the gage should incorporate features facilitating their calibration. The same thing can be done on a CMM without a whole lot of fuss and feathers. Damn new fangled gadgets anyway.

                    As a matter of fact I designed a set of A11 gages for a Landis roll grinder headstock. I made the gages to match faces with the tapers having zero radial play. The female product taper was gaged by recording radial movement of the male gage assembled on the part. I could not for the life of me think of a better way to do it and still have the gages self check each other's taper - the male gage being the only one measureable outside of a metrology lab.
                    Last edited by Forrest Addy; 12-12-2010, 01:10 PM.

                    Comment


                    • #11
                      It would be helpful if the OP indicated how much runout he is
                      getting. As Forest has posted in the past 3J chuck runouts
                      of less than 6 thou or so are excellent and par. I have found
                      with my D1-4 that it makes a difference which post on the
                      chuck goes in which hole, it also helps to torque the cams
                      in the right direction (early in my experience I could not keep
                      track CW or CCW to turn the cams). Try all three holes,
                      rotating the chuck 120D, and lightly mark the spindle to index
                      it first and see if you get better indexing at one hole. The
                      amount of torque placed on the cams can influence runout as
                      well. Not to go into cam pin positioning, dings, debris etc.
                      Determining if the spindle is off is a high art as you are measuring
                      in the sub thou range.
                      Steve

                      Comment


                      • #12
                        At the risk of being a party spoiler here, I have to say I disagree with many of the comments made in this thread. Let me explain and please approach this with out pre-concieved ideas

                        In any tooling or production setup, you can only have ONE boss per axis. This is the cardinal rule for jig and fixture design.
                        You can see this easily with the common knowledge that 2 points make a straight line. So when you make a fixture , you use two points to hold a part that you want straight/parrallel to your desired position. you don't use 3 or 4.
                        For a "plane" you would need 3 points, because we know that 3 points that are not collinear (not in allignment) make a plane
                        Why am I saying this ?
                        Because you cannot have a taper and a plane function for the same purpose !
                        They are in direct conflict with each other
                        Do you know of any tapered collet that that has a stop ?
                        Now look at the design for a D1-5 taper
                        Note the VERY specific Taper (7 Deg 7.5 min = 3"/Ft combined ang)and Specific Diameter 2.5005.
                        The Tolerance is +.0005, thats it !
                        Note the the spindle nose flat is not a dimensioned distance from the sharp line of the taper. Note also that the face to the pilot nose is a fraction ( 7/16 in the real specs)
                        Why would this be critical if the specification is a fraction ?
                        What this means is that there is no concern by the machine tool industry for the "space" that some here are concerned over.
                        When the chuck is drawn up evenly on the pins, the chuck will be proud, of the face. I have seen this in most all the shops I have been associated with and the pins visible as well.
                        As mentioned earlier, there is one boss, and it is the taper , as it is the only critically diimension component.
                        Pulling flush to the face is nice, but not specified, which means it cannot always be expected
                        Rich
                        Last edited by Rich Carlstedt; 12-12-2010, 04:12 PM.
                        Green Bay, WI

                        Comment


                        • #13
                          I guess you haven't heard of HSK toolholders.

                          Phil

                          Originally posted by Rich Carlstedt
                          In any tooling or production setup, you can only have ONE boss per axis. This is the cardinal rule for jig and fixture design.
                          Rich

                          Comment


                          • #14
                            Originally posted by Mcgyver
                            clear on paper, difficult to measure or to do. Most spindles will have a relief there, or should, so the intersection of cone and face is a theoretical place not actual. Even if it were actual, any radius on the tool or grinding wheel that cut it will wreck the measurement....and what measurement tool do you get in there that is reliable to 1/2 a thou?

                            When I've done it I've used a female i trust (not many of those lol), turned the taper first, then start kissing off thou's on face using feeler gauges to know how much to take off. I didn't know until Forrest's post that it was supposed to be a thou but that's probably where it ended up.
                            Ok, good point, the taper on the adapters I have been making for the rotary tables go back to the face. The taper in the pic that the OP posted looks to me like it goes beyond the chuck face and the relief is cut outside of this. I could be wrong and these old tired eyes are getting worse than I thought.

                            In retrospec 2.5005 is what I shoot for but I use my 8" Bison chuck for the test fit and carefully take off .001 at a time until it does have just about a .001 proud of the face. The pins then pull it tight to the face of the adapter.

                            If the OP is talking about this kind of a fit then I don't think there is a problem with the taper.
                            Last edited by DeereGuy; 12-12-2010, 06:20 PM.
                            Bob
                            Pics of shop and some projects
                            http://s2.photobucket.com/albums/y39...achine%20Shop/

                            Comment


                            • #15
                              Phil, I think our discussions have focused on rigid tool systems.
                              While HSK is a supreme toolholder, it is designed to rely on the flange for support and the tapers (very thin wall) are designed to " flex" until the flanges interface. The extreme precision of these holders make them unique compared to all other systems

                              I quote a popular discription for those unfamiliar with HSK

                              "As the collet segments engage, the pullback force causes the shank to deform slightly, pulling the shank further into the receiver until contact is made between the flange and the receiver's face. The fact that the two tapers are in intimate contact due to the elastic deformation of the shank ensures the positioning accuracy and repeatability of the joint, both radially and axially, to 0.0001".

                              http://www.hskworld.com/articles/hsk...sk_secrets.htm

                              The HSK components are designed and ground to have a .00025 "( .006mm) clearance flange to flange when the flexible tapers (non-solid) engage, then added force is used to flex the taper and bring the flanges together.
                              In this case, the boss is the flange as further loading produces no change.

                              A D1-4 taper spindle, or chuck is not designed to flex, but considered a rigid tool system
                              Thanks for bringing it up.
                              Rich
                              Green Bay, WI

                              Comment

                              Working...
                              X