Rotary Tilt Compensation

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

  • Rotary Tilt Compensation

    What's the best way to compensate for tilt when an overly long part is chucked in a horizontal rotary table?

  • #2
    Crank it around until its not tilted?

    Not sure I understand what you are asking.

    Comment


    • #3
      Let's say your rotary table has teeth that stick 3/4" from the face. Logic holds that that means you can grab 3/4" of the part.
      Let's say your part is 13 inches long and 12 lbs, and gravity and the like causes it to tilt.

      EDIT:
      No matter how hard the rotary is cranked, it will never hold the part perfectly straight

      Comment


      • #4
        Could you use/make a V block on an adjustable stand to support your part? Or rig up some centers?

        Comment


        • InspectorJester
          InspectorJester commented
          Editing a comment
          I was thinking the same thing, but I am not sure it is feasible given the setup

      • #5
        The term compensation is where I'm a bit lost. Compensation is taking error into account, not fixing it. So it sounds like you aren't looking to correct the physical tilt, but to compensate for it. Doesn't the alignment take care of that, or am I missing something?
        PC-DMIS 2016.0 SP8

        Jeff

        Comment


        • #6
          If you align on the part, and then rotate the table, I would expect PC-DMIS to do the right thing - are you saying it doesn't (I don't have access to a rotary table)?
          AndersI
          SW support - Hexagon Metrology Nordic AB

          Comment


          • #7
            Schrocknroll : Yes; by "compensation" I mean that I don't believe there's an adequate way to fix the tilt, so I have to account for it. I am searching for the "best method".

            AndersI I would as well. Perhaps my alignment is not constructed accurately, or there is something wrong with the calibration, but I'm looking for a "standard procedure" for accounting for something like that.

            If, when everything is set up correctly, even if there is massive tilt, the probe will keep center regardless of the angle of the rotary, then I need to take a look at my alignment. If the alignment is sound, I need to investigate the calibration of the rotary

            Comment


            • #8
              if you are talking about the part 'flopping' as you rotate the rotator due to gravity and weight (like a hot dog on a stick), then you may have to re-align every time you rotate. I don't think any alignment or calibration of the rotary will fix the 'angle of the dangle' that is caused by weight & gravity as you spin the part.
              sigpic
              Originally posted by AndersI
              I've got one from September 2006 (bug ticket) which has finally been fixed in 2013.

              Comment


              • louisd
                louisd commented
                Editing a comment
                lol "Angle of the dangle" that made my week!

              • InspectorJester
                InspectorJester commented
                Editing a comment
                That's my common practice; realigning after each rotation. It adds a buttload of time to the program that I don't want if I don't need though.
                EDIT:
                I'm not sure if this is what you meant, but the part doesn't move once it's in the rotary, it just sits in at a weird angle that gets weirder as I turn

            • #9
              What about tooling a centering-spike at the same height as the rotary, that can retain the part along the axis, instead of rigging a v-block? Just make sure it's versatile enough to hold whatever length your parts are...
              If most of your product is turned, I'm sure the machinists have centers on the ends of all your parts anyways, yeah?

              Comment


              • InspectorJester
                InspectorJester commented
                Editing a comment
                Nope. Most of them have threads on the end, so if there was a center, it would be removed. Otherwise they only center on one end or something; I'm not a machinist, and will try not to pretend I know what I'm talking about :P
                If they are threaded, I could still make it work, no?

            • #10
              Originally posted by InspectorJester View Post
              What's the best way to compensate for tilt when an overly long part is chucked in a horizontal rotary table?
              Root Cause: insufficient holding force with current fixturing setup chuck jaw configuration.
              Corrective Action: upgrade fixturing setup with better chuck jaws.
              /thread

              Comment


              • louisd
                louisd commented
                Editing a comment
                Hi Ego Murphy, welcome to the forum. Good insight.
                There's often more than one way to solve a relatively simple problem, so your whole mic-dropping /thread is rude and closed-minded.

                Root cause is actually the effects of GRAVITY. Add a center on opposing end of part, or rotate/fixture the rotary to orient the part vertically. All loads are evenly distributed when gravity isn't pulling the part down while on its side.

                Also, in 99.9% of real world cases, a bigger rotary would get promptly shot-down for budget constraints.

            • #11
              I was thinking about something like a counter spindle at the right height... Same idea that louisd , for OD.

              Comment


              • #12
                A. | A bigger rotary was already shut down, due to budget constraints. Even some slightly longer jaws couldn't be acquired, unfortunately.

                B. | I am not entirely sure it is necessarily the fault of gravity. At least not entirely.
                The reason being, it doesn't move when it's in the chuck, it just sits in angled; when I rotate to 180, for example, it angles the opposite way. Sometimes, this action causes the probe to miss the part.
                Perhaps my alignment does not accurately compensate for it, but I was under the impression it does.

                C. | I am curious of this counter spindle idea. Given the scenario in B., how would this not damage the part?
                Perhaps I just do not understand the setup.

                Thanks all for your replies

                Comment


                • #13
                  Originally posted by InspectorJester View Post
                  A. | A bigger rotary was already shut down, due to budget constraints. Even some slightly longer jaws couldn't be acquired, unfortunately.

                  B. | I am not entirely sure it is necessarily the fault of gravity. At least not entirely.
                  The reason being, it doesn't move when it's in the chuck, it just sits in angled; when I rotate to 180, for example, it angles the opposite way. Sometimes, this action causes the probe to miss the part.
                  Perhaps my alignment does not accurately compensate for it, but I was under the impression it does.

                  C. | I am curious of this counter spindle idea. Given the scenario in B., how would this not damage the part?
                  Perhaps I just do not understand the setup.

                  Thanks all for your replies
                  In this case, I think I would set the part on the counter spindle (or set this one on the part ?) and then fixture the part on the rotary.
                  Maybe some settings with a dial comparator before fixturing would be usefull.
                  But it's time consuming, so not adapted to a serial control.

                  Comment


                  • #14
                    Originally posted by InspectorJester View Post
                    A. | A bigger rotary was already shut down, due to budget constraints. Even some slightly longer jaws couldn't be acquired, unfortunately.

                    B. | I am not entirely sure it is necessarily the fault of gravity. At least not entirely.
                    The reason being, it doesn't move when it's in the chuck, it just sits in angled; when I rotate to 180, for example, it angles the opposite way. Sometimes, this action causes the probe to miss the part.
                    Perhaps my alignment does not accurately compensate for it, but I was under the impression it does.
                    (pics up mic, dusts it off, apologizes to LouisD)
                    If you are hitting/missing one end, then your alignment is insufficient.
                    I fully support having a "helper alignment" that levels to an axis that runs through the entire part, regardless of flimsy datum scheme.
                    You can also use little local alignments to account for a warped part.

                    Originally posted by InspectorJester View Post
                    Thanks all for your replies
                    You are welcome, I am please to help out

                    Comment


                    • InspectorJester
                      InspectorJester commented
                      Editing a comment
                      I tend to use the local alignments strategy when necessary, but sometimes there is a different tilt from part to part, so the alignment needs to be adjusted; hence why I'd like to better compensate for it and [theoretically] not need them

                  • #15
                    Originally posted by InspectorJester View Post
                    B. | I am not entirely sure it is necessarily the fault of gravity. At least not entirely.
                    The reason being, it doesn't move when it's in the chuck, it just sits in angled; when I rotate to 180, for example, it angles the opposite way. Sometimes, this action causes the probe to miss the part.
                    Perhaps my alignment does not accurately compensate for it, but I was under the impression it does.
                    Are *you* or *PC-DMIS* rotating the table? I.e. is PC-DMIS aware of the rotation? Do your hits work in the first position before rotating?

                    AndersI
                    SW support - Hexagon Metrology Nordic AB

                    Comment

                    Related Topics

                    Collapse

                    Working...
                    X