Profile Gauge Check

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

  • Profile Gauge Check

    I have a part that has a profile called out to ABC and a tolerance of +,- 3mm but the part is fails when measuring on CMM. So here at work an Engineer created a "Profile Gauge" which is a piece of metal with the outline of the part cut out and off set edges which is suppose to represent the 3mm tolerance.

    I don't think this Profile Gauge works to well because the same part I measured on the CMM (the part that failed), went right threw his Profile Gauge, which to him means the part passes profile, But my argument was that he wasn't checking Profile to the Datum structure called out on the print which is ABC, I feel like he is measuring profile free of datum structure.

    do you guys think a "Profile Gauge' is a valid method? Have you guys used Profile Gauges? if so how do you verify them on the CMM? Thanks in Advance

    update- I think this called a Profile Template.
    Last edited by PacMan23; 10-19-2017, 04:20 PM.

  • #2
    A drawing/sketch of the print would help.
    Are we talking profile of a line, or surface?
    Are the datums internal and the perimeter of the part being controlled via profile (all around)?
    Fundamentally speaking, you can functional gage any feature(s) of size. Yes a functional gage could validate acceptance of a profile FCF. As far as how complicated the functional gage needs to be, is a different story.

    Comment


    • PacMan23
      PacMan23 commented
      Editing a comment
      I'll try and draw something up quickly

      the Dimension is Profile of a Surface

  • #3
    side view.PNG

    Comment


    • #4
      so datum A is the Plane and datum's B and C are 2 holes that lay on Datum A. and Dimensioning the surface called out to ABC


      this side view of the part is exactly what the Template looks like. so Just imagine a piece of metal with this outline cut out

      Comment


      • #5
        Have you tried checking the part with a surface plate setup?

        Clamp -A- to an angle plate, level off of on -B- and -C- ...........

        Comment


        • PacMan23
          PacMan23 commented
          Editing a comment
          Yes and you can visually see that the part is failing

      • #6
        I can't see a profile template as a viable option, if they really wanted to use something like that they'd need 2, 1 for the high end and one for the low end. Even then your CMM is going to be more accurate I've had plenty of parts that fail CMM but pass gage.

        Comment


        • #7
          I have had this discussion so many times where I work.

          As you say, they are basically checking the profile and best fitting it. Where as you are checking the profile to datums, you are always going to get different results.

          If you were to change your dimension in pcdmis to just profile and best fit you may well pass the part. But if the drawing states profile to datums then that's what it should be.

          Are there any other methods you can check the part? laying it on the CMM and using a height gauge or similar? I know its not easy with a profile callout.

          I am going through this process right now with our sister site. They have been setting a part up with hard gauges and every part passes, they then had there CMM guy measure the same parts and they were also good. However they have problems with fit and function. I have just measured 5 parts and found them all to be out of spec. The reason this was happening is both the gauge and the CMM guy down there checked the part to just profile and not datums.

          Unfortunately now I have done this they are sending 120 parts up for measurement, just my luck.

          Comment


          • PacMan23
            PacMan23 commented
            Editing a comment
            hahaha Hey you guys have the same system we do, "If you do your job good enough then you get to do other peoples job"

          • PacMan23
            PacMan23 commented
            Editing a comment
            yes and it seems they have had that same type of issue before, where parts failed but passed gauge checks and sent those parts out only to be rejected. And again today they are trying to do that with this part.

            I know overall profile will pass but to the print Callout of ABC it is out of Spec, But if they can verify this template is within tolerance then they think they can also say the part is within tolerance. Its like comparing Oranges to Tangerines, they're kinda not the same.

        • #8
          The problem with this is similar to the problem people see when they switch the order of features on a perpendicularity callout.

          Datum A is a very thin surface compared to the surface being checked. If you were to lay the part on the surface in question, and measure from it to the center of B and C, you would get a wildly different number than the methods described above of leveling to A and then checking the distances from B/C to the surface near the bend and far away from the bend.

          This will be hugely exaggerated by changing the angle between A and the surface being checked.

          A simple sanity test would be to check the angle from A to that surface. +3mm at 100mm length (no idea how tall that surface actually is away from A) would be 1.718 degrees. So, if the surface was only off in orientation, then it could only bend by +/- 1.718 degrees. If that surface is too close, or too far, from B/C, then even less angle would be required to cause an out of tolerance condition.

          Another simple check, after checking orientation, would be to dimension out the actual value of the surface from B/C. If it is perp to A, then it should be x.xx(mm) away from B/C at all points. So, list out the Z value of those points. They should all be at x.xx +/- 3mm.

          The only way that profile gage could work is if it contacted A and pinned B/C. Anything else is junk.
          Last edited by VinniUSMC; 10-20-2017, 06:41 AM.
          "This is my word... and as such is beyond contestation."

          Comment


          • PacMan23
            PacMan23 commented
            Editing a comment
            I know for a fact that the part is failing, I have a magnetic table top where I can pin Datum A, flat and then to visually show how and where the part is failing profile I just lay a stop block next to the surface being dimensioned and it makes contact at the top of the surface being dimensioned but has a large gap at the bottom.

          • PacMan23
            PacMan23 commented
            Editing a comment
            thanks for the input, I agree you would have to control A,B,C before you could do anything else

        • #9
          Maybe the print is controlling the part incorrectly?
          It makes logical sense to datum the surface from which the nuts are on, but if that current datum A plane is 1/4th the amount of surface that the profile measurement is on, it will project and compound any angularity error between surfaces, by a factor of 4.
          Where are the basic dimensions (if any), or is the profile surface controlled 90° from GD&T rule 1 (implied orthogonal)?
          Last edited by louisd; 10-20-2017, 10:57 AM.

          Comment


          • #10
            yes there are basic dimension but I eliminated them from the post, I agree that the print makes it Very difficult to profile that surface with that type of datum scheme, and I have expressed this with Engineering, I told them that the part Looks relatively simple not to many holes etc. but what makes it so difficult is the Datum Structure. this is common with parts from this new customer- And they wont change the print because they identify their Datums as the surface or holes that first make contact with the mating parts, which makes sense but I don't think they realize the difficulty of it.

            Comment

            Related Topics

            Collapse

            Working...
            X