Tooling Balls

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

  • Tooling Balls

    Howdy:

    I work for a company and we use dedicated fixtures for part layouts...

    We recently received some fixtures that were built on a CMM platen...

    They were then sent to our facility and transferred to stackable cradles with each fixture sitting on top of adjuster nuts...

    The tooling balls on some of these fixtures are now wonky...

    Each fixture has a left and right hand part and uses tooling ball number three for both hands of the part (iterative alignment)...

    We were having to use shims for the number three tooling ball to get it in spec before we ran the parts...

    We suspect the cradles have warped the fixtures but the powers that be won't do anything about it...

    Now they have told us to stop adjusting the tooling balls and just run the alignment with the tooling balls out of spec (they figure it won't cause that much error)...

    I remember seeing in posts on one forum or another that we can actually do another alignment in order to transfer the tooling ball error to the part but I'm not sure how to do it and neither does my partner in crime...

    Your thoughts on this would be appreciated...

    Running an alignment with the tooling balls out of spec is not the solution...

    Help?

    T

  • #2
    You can set up an iterative alignment from the part's datum locators, then measure the tooling balls to get your amount of error if that's what looking to do.
    Xcel & MicroVal Pfx & Global 37mr4 thru 2012mr1sp3
    Contura Calypso 5.4

    Lord, keep Your arm around my shoulder and Your hand over my mouth. Amen.

    Comment


    • #3
      Originally posted by dwade View Post
      You can set up an iterative alignment from the part's datum locators, then measure the tooling balls to get your amount of error if that's what looking to do.

      Keep an eye on the XYZ relatrionship of the part's datum locator's to each other as-well, in case they've been affected.
      sigpic

      Comment


      • #4
        I would just do what they said, when there is a problem I would tell them it's probably because of the tooling ball. It's a battle not worth fighting, just do as your told. You could also check a part the way the tooling balls are , and recheck the same part with the tooling ball adjusted and show them the difference between the 2 reports, there may not be that big of a difference.
        sigpic


        Was
        Pc-dmis 3.5 MR1 B & S 2009 MR1

        Now
        2010 MR3

        Comment


        • #5
          Originally posted by dwade View Post
          You can set up an iterative alignment from the part's datum locators, then measure the tooling balls to get your amount of error if that's what looking to do.
          There ya go...
          RFS Means Really Fussy Stuff

          When all you have is a hammer - everything looks like a nail....
          sigpic

          Comment


          • #6
            Originally posted by Treeant View Post
            Howdy:

            I work for a company and we use dedicated fixtures for part layouts...

            We recently received some fixtures that were built on a CMM platen...

            They were then sent to our facility and transferred to stackable cradles with each fixture sitting on top of adjuster nuts...

            The tooling balls on some of these fixtures are now wonky...

            Each fixture has a left and right hand part and uses tooling ball number three for both hands of the part (iterative alignment)...

            We were having to use shims for the number three tooling ball to get it in spec before we ran the parts...

            We suspect the cradles have warped the fixtures but the powers that be won't do anything about it...

            Now they have told us to stop adjusting the tooling balls and just run the alignment with the tooling balls out of spec (they figure it won't cause that much error)...

            I remember seeing in posts on one forum or another that we can actually do another alignment in order to transfer the tooling ball error to the part but I'm not sure how to do it and neither does my partner in crime...

            Your thoughts on this would be appreciated...

            Running an alignment with the tooling balls out of spec is not the solution...

            Help?

            T
            I've been trying to figure this ine out for a bit and I still don't get it. There should be 3 tooling balls for EACH part to set the alignment. You are doing ITERATIVE alignments. Now, if the 3 tooling balls are in a base, and you use those 3 balls for the alignment, HOW is it possible that you would need to SHIM one of them to get it into spec? That dieection would be the FIRST thing the iterative alignment solves and the 'up/down' direction, so to speak, would be PERFECT for all 3 balls. Are you NOT using them for the alignment? Are you using something else for the alignment, THEN checking the tooling balls? If you put 3 t-balls in a fixture base, but then have to shim one to get it into spec (AFTER the alignment), then you must have bent that base VISABLE for a 'shim' move to fix the t-ball location and then you would have GROSSLY HUGE deviations in the other axis as well. I just don't get it. Anyone else get it?
            sigpic
            Originally posted by AndersI
            I've got one from September 2006 (bug ticket) which has finally been fixed in 2013.

            Comment


            • #7
              Matthew,

              I'm with you. I also do not understand.

              As others have stated above, perform an alignment on the fixture, based upon the datum locations for the part. Then, go back and measure the tooling balls to see how far off they are.

              If the fixture has been bent, there is no way on God's green earth that you can perform an iterative alignment on the tooling balls, and have the part measure within specification.

              Blessings,
              ZydecoPete
              sigpic

              Comment


              • #8
                Originally posted by Matthew D. Hoedeman View Post
                I've been trying to figure this ine out for a bit and I still don't get it. There should be 3 tooling balls for EACH part to set the alignment. You are doing ITERATIVE alignments. Now, if the 3 tooling balls are in a base, and you use those 3 balls for the alignment, HOW is it possible that you would need to SHIM one of them to get it into spec? That dieection would be the FIRST thing the iterative alignment solves and the 'up/down' direction, so to speak, would be PERFECT for all 3 balls. Are you NOT using them for the alignment? Are you using something else for the alignment, THEN checking the tooling balls? If you put 3 t-balls in a fixture base, but then have to shim one to get it into spec (AFTER the alignment), then you must have bent that base VISABLE for a 'shim' move to fix the t-ball location and then you would have GROSSLY HUGE deviations in the other axis as well. I just don't get it. Anyone else get it?
                Originally posted by Peter Warcholyk View Post
                Matthew,

                I'm with you. I also do not understand.

                As others have stated above, perform an alignment on the fixture, based upon the datum locations for the part. Then, go back and measure the tooling balls to see how far off they are.

                If the fixture has been bent, there is no way on God's green earth that you can perform an iterative alignment on the tooling balls, and have the part measure within specification.

                Blessings,
                ZydecoPete
                But for a 'shim' move to fix a t-ball USED in an iterative alignment? The ONLY way I can see that a shim would fix it would be if the alignment WAS NOT coming off the t-balls to begin with. The 'height' is the FIRST thing solved and no matter HOW much you shim it, it will not change! IT will ALWAYS be the same. He MUST be using something else for his alignment and then goes back and checks the t-balls. If the fixture is made/certified with the t-balls, USE them, and ONLY them to do the iterative alignment, do not use the base, then check the balls, use the balls, then check the part locators.
                sigpic
                Originally posted by AndersI
                I've got one from September 2006 (bug ticket) which has finally been fixed in 2013.

                Comment


                • #9
                  Originally posted by Matthew D. Hoedeman View Post
                  I've been trying to figure this ine out for a bit and I still don't get it. There should be 3 tooling balls for EACH part to set the alignment. You are doing ITERATIVE alignments. Now, if the 3 tooling balls are in a base, and you use those 3 balls for the alignment, HOW is it possible that you would need to SHIM one of them to get it into spec? That dieection would be the FIRST thing the iterative alignment solves and the 'up/down' direction, so to speak, would be PERFECT for all 3 balls. Are you NOT using them for the alignment? Are you using something else for the alignment, THEN checking the tooling balls? If you put 3 t-balls in a fixture base, but then have to shim one to get it into spec (AFTER the alignment), then you must have bent that base VISABLE for a 'shim' move to fix the t-ball location and then you would have GROSSLY HUGE deviations in the other axis as well. I just don't get it. Anyone else get it?
                  Agree you also would need to re verify fixture shim datums you can't just change offsets on tooling balls and make it all better!!1
                  sigpic
                  3.7 MR4

                  Comment


                  • #10
                    Originally posted by Matthew D. Hoedeman View Post
                    Now, if the 3 tooling balls are in a base, and you use those 3 balls for the alignment, HOW is it possible that you would need to SHIM one of them to get it into spec?
                    What I got out of his explanation was that the base-plates are distorted, thus changing the relationship of the tooling balls to the part datum holding features........................and his boss, or whomever, doesn't seem to care and says use it anyway.

                    This being the case, it is possible that the part datum holding features still maintain a relationship to each other that makes them sort-of usable, albeit highly undesirable in my opinion.

                    I've been down that road before. It's called real-life dumb bosses, or some sort of stupidity.
                    sigpic

                    Comment


                    • #11
                      Originally posted by Matthew D. Hoedeman View Post
                      But for a 'shim' move to fix a t-ball USED in an iterative alignment? The ONLY way I can see that a shim would fix it would be if the alignment WAS NOT coming off the t-balls to begin with. The 'height' is the FIRST thing solved and no matter HOW much you shim it, it will not change! IT will ALWAYS be the same. He MUST be using something else for his alignment and then goes back and checks the t-balls. If the fixture is made/certified with the t-balls, USE them, and ONLY them to do the iterative alignment, do not use the base, then check the balls, use the balls, then check the part locators.
                      Quote(Each fixture has a left and right hand part and uses tooling ball number three for both hands of the part (iterative alignment)...
                      )


                      maybe 5 tooling balls would show that the shared # 3 is out of wack.

                      Comment


                      • #12
                        Isn't that a part of our business?

                        "Dumb bosses", "Skilled engineers", etc.

                        After all these years in the quality field, especially as it pertains to programming and running a cmm.

                        If my numbers show a part as being "good", I've done a great job.

                        If the numbers are "not so good", it's "Are you sure that you checked it correctly?"

                        This unfortunately, is a part of the great burden we must bear!!

                        Blessings,
                        ZydecoPete
                        sigpic

                        Comment


                        • #13
                          Clarity...

                          Morning All:

                          There are five tooling balls affixed to the base plate...

                          The part is in CAR position the X and Z axes are the base plate axes, the Y-axis is the vertical axis on the CMM (left hand rule tipped up)

                          The #3 tooling ball is used by both parts for each of their respective iterative alignments...

                          I pick up the three tooling balls for the left hand part (for example) it then goes into DCC and checks the balls...

                          It returns a value for the number three tooling ball that is beyond the 0.05mm tolerance that we attribute for it's positional location (usually less than 0.15mm deviation)...

                          If I were to then pick up the tooling balls for the right hand part, it would go into DCC mode and return a value of 30 microns beyond nominal for that number three tooling ball (within tolerance)...

                          Prior to putting this fixture on a stackable cradle, both parts would show tooling ball #3 as being within tolerance on both sides of the part after PC-DMIS had done the iterative alignment in DCC mode...

                          I would put a shim under one edge of the number three tooling ball (in order to tip the center of the ball in the Z-axis) on the side that was out of spec and the iterative alignment would change the returned values for all three tooling balls but they would all be within tolerance and then we would proceed to inspect the part...

                          The powers that be then told us to stop this practice as it was sometimes time consuming to find the sweet spot where the shim would return values within tolerance for all three tooling balls...

                          Now we are using the fixture with the one side's tooling ball #3 out of spec (if the customer knew we were doing this they would probably have a bird)...

                          As I said before, I was told long ago that some people do another iterative alignment after the initial aligment to get better results but neither my partner nor myself have ever done this...

                          Nor do we know if that would be a feasible solution; however, it sounds like it might be a quick solution...

                          I'm just trying to protect the customer...

                          We both think that the cradle has somehow bowed the fixture and this is what's causing tooling ball #3 to be out of position as nothing else has changed...

                          Help?

                          Comment


                          • #14
                            Maybe the fixture has more that 3 tooling balls, Maybe 6.
                            If thats the case one or more of them could certainly be "wonky".
                            Shimming probably takes some time and that is what is giving your boss (or his boss, or some dam boss) grief. "The dam parts are stuck in inspection" I can just hear it.

                            I would follow the above advice about checking without shim and then checking with shim and see what the differences are.

                            (EDIT) I see you were typing at the same time I was.... hang in there.
                            Lately, it occurs to me
                            What a long, strange trip it's been.

                            2017 R1 (Offline programming)

                            Comment


                            • #15
                              Have you tried to verify the datum locators on the fixtures to the tooling balls. The tooling ball locations are reference only, generally you would like to see them within 0.05 but due to environmental conditions between the inspection lab and your lab, uncertainty, calibrations and just about everything else in the world these numbers just are not achievable. I would recommend verifing the fixture, either in house or send it out to a lab. Drop me an email and I will do my best to help you through the verification.
                              Dan
                              [email protected]

                              Comment

                              Related Topics

                              Collapse

                              Working...
                              X