No announcement yet.

Dialing in a shaft...

  • Filter
  • Time
  • Show
Clear All
new posts

  • Dialing in a shaft...

    It's been a while since I did much serious machining and I ran into an issue today while setting up to machine the end of a shaft that had been built up with weld. Both ends of the shaft are centre-drilled so I grabbed one end in the 4-jaw and supported the other with a live centre. I dialed the chuck end in to within .001" and then put the indicator on a machined section of shaft on the tailstock end. Runout was about .004". Assuming that the centre hole was out a bit I set up the steady rest and dialed the shaft in till it was running within .001". I then ground a piece of HSS and used it to true up the centre on the end of the shaft. However, when I loosened the steady and pushed the live centre back into the shaft I still had about .004" runout.

    This lead me to think that the centre hole in the shaft was true to begin with and all I did was clean it up, not change it. Having ruled out the centre hole I thought that perhaps the live centre itself was out. I tried a second live centre with roughly the same results so that points to the tailstock being out of alignment.

    In conjunction with the idea of something being physically out of alignment the other thing that started to bug me was the technique of getting the shaft centreline aligned to the centreline axis of the lathe. When I dialed in the shaft on the tailstock end I used a single dial indicator reading on the side of the shaft. When the indicator was reading less than .001" runout was the shaft centreline really that close to the axis of the lathe? Should I have used a second indicator on the top of the shaft to ensure that I was aligned both vertically and horizontally? I've done this kind of stuff before and should know but it's been awhile and I seem to be having a brain f*rt which won't let me wrap my head around this...
    Just one project too many--that's what finally got him...

  • #2
    Whoa horses!!

    An out of line tailstock does not effect rotational runout.

    You don't say where the welded area is in relation to either center hole? Or where the welded area is in relation to the spots where the indicator was running when you dialed the shaft in originally?

    I can guarantee that the welding warped the shaft to some extent, usually rendering any existing center holes useless as is. You did use the proper technique to true up the center hole, trying to do it with a center drill is usually an exercise in futility as the drill will only follow the existing hole. A less time consuming method is to use pie shaped shims between the center point and center hole, though it can be hairy to ensure that they don't fall out at the wrong time.



    • #3
      If you dialled it in at the chuck, then saw some wobble at the other end, that's pretty normal. You could then flex it around a bit to see if you can center it- still only the chuck holding it. If you can make the wobble disappear, then the shaft is aligned to the lathe axis.

      Chances are there would still be some wobble. You could set up the steady so each jaw just touches at one point as the chuck rotates. You would see a gap between the shaft and each jaw grow to a maximum, then come around to barely touching at one point. Each jaw would then be adjusted inwards by exactly half of the maximum gap you saw. If that goes well, then all the jaws are touching, the wobble is gone, and the shaft is true to the lathe axis. You could take an indicator reading from the side of the shaft at both ends, then adjust the steady jaws to bring the readings to be the same. The shaft might end up being slightly high or low, but it will be parallel to the axis in the fore and aft direction, which is what's important to be able to turn it without creating a taper. A very slight up or down misalignment isn't going to affect this to any degree that you could measure.

      At this point you should be able to turn the shaft without it walking out of the chuck, and then you should be able to see whether the center divot is really centered. I'm assuming it didn't warp out of place because of the build-up welding- maybe a fair assumption, but I don't know whether the welding warped the end of the shaft. I am assuming that the welding is close to the end, so it shouldn't throw the dimple off by much- but I don't know.

      If you have determined that the dimple runs true, then you can see whether the tailstock center is on center- theoretically it should nest right into the dimple without anything moving. If the center wants to push the shaft one way or the other, then it's the tailstock that's out.

      I'm a bit confused- you said runout, and talked about adjusting it out- did you mean wobble, or a radial misalignment shown by taking readings from the side of the shaft at both ends? The steady is going to take the wobble out, but you could still have this radial misalignment at the tailstock. Careful adjustment of the steady can take this out as well, as I talked about above- I'm just not sure what you really mean-

      Either this helps, or it's more confusing-

      Oh, I may have missed the obvious- is the shaft straight? If it isn't, you are going to have your hands full of worms-
      Last edited by darryl; 04-11-2013, 04:16 AM.
      I seldom do anything within the scope of logical reason and calculated cost/benefit, etc- I'm following my passion-


      • #4
        When fixing a center, I just put the drill holder to the QCTP, chuck a center drill in so that the lip closest to me is about at center height and use it as a boring tool with very very light hand feed to correct the center. Just yesterday did that when an existing shaft needed some truing up to the next size down on some bearing spots.

        But yeah, would need information on where exactly is the weld, etc etc, a photo would be best.
        Amount of experience is in direct proportion to the value of broken equipment.


        • #5
          What Dave said. The shaft is probably warped from welding and chucking in the four jaw is forcing the tailstock end to run out. The best method for this type of work is to turn between centers.
          Jim H.