Help with Datum structure.

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

  • Help with Datum structure.

    How would I go about setting up this Datum Structure? cap2.PNGcap1.PNG

  • #2
    Read pnt over/in -A-. With sample hits, measure -A- as a cylinder. Make the cyl X&Y&Z origin. Then, throw a line for alignment through the cyl.

    Then..Level to the cyl, rotate to the line, make -A- XYZ origin. Then, measure -B- as a cyl with sample hits. Then, draw a line from -A- to -B-. Then put a point in the middle of said line.

    Then... level to -A-, rotate to the line between cyls, then make your pnt X&Y 0 and -A- Z0.

    Report position.
    SF7107(PCD), SF454(PCD), 152614(Quindos), 9159(Quindos), 7107(Quindos), B&S Manual, M&M Gear Checker

    Comment


    • PacMan23
      PacMan23 commented
      Editing a comment
      thanks, I'll see if I can make this work.

  • #3
    Looking at this, all six degrees of freedom cannot be restrained by the datums shown. If the part were to be loaded onto a CMM with the A and B accessible in Z axis, nothing is shown to control location of this information in the Z axis.
    A = level and 2 origin or level and 1 origin
    B = rotation or rotation and 1 origin
    C = Nothing left to control

    Atleast thats what I see looking at the sketches

    Comment


    • #4
      Looks incomplete to me as R2ah1ze1l stated above. Nothing controls translation in Z.

      Comment


      • #5
        I believe what the engineer might be looking for here, from what I see, Please correct me if I'm wrong. Datum -A- is the center thru hole, and you have 2 weld nuts, one on each side of the plate, basically making sure the weld nuts are centered to Datum -A-, same goes for Datum -B- 2 weld nuts on each side, except it needs to be Parallel to -A- because welding is going to warp it. and -C- just needs to be Perpendicular to both holes. Just looks like an assembly drawing. I would just use the plate to origin Z. perpendicular to the hole.
        Last edited by KIRBSTER269; 12-06-2017, 11:42 AM.
        (In Memory of my Loving wife, "Ronda" who I lost March 7, 2016. I love you baby.)
        They say "Nobody's Perfect." I must be Nobody.

        Comment


        • PacMan23
          PacMan23 commented
          Editing a comment
          So what do you mean when you say that its an assembly drawing?? does that mean I should create the alignment with the Datums the print calls out or does that mean I could create my own alignment and just make sure everything is where it needs to be?

        • KIRBSTER269
          KIRBSTER269 commented
          Editing a comment
          Level the Plate and origin Z, origin X and Y to datum Hole -A-, Rotate to Datum -B- from -A-, for the alignment, create different cylinders, circle with thread pitch on each nut separately and constructed together, and report Parallel between the 2 holes, even though its not called, position from one hole to the other. Making sure that the nuts are coaxially mounted together. Perpendicularity like the callout, and even Perpendicularity of the plate itself back to -C-, but also you could do as Andersl suggested. And compare the 2 alignments you are already creating both alignments. create the other alignment without creating more features.Use the ones you got. Your tolerance is pretty sloppy so you have a lot to work with. PacMan23
          Last edited by KIRBSTER269; 12-06-2017, 04:57 PM.

      • #6
        It says TWO COAXIAL WELDNUTS - doesn't that mean that [A] is a cylinder/line through the whole part, from left to right? Locking two directions and two coordinates. [B] is another line through it all, locking the rotation (not by its direction, but by its location). [C] is the mid plane of the two ends, locking the final coordinate. Or...?


        AndersI
        SW support - Hexagon Metrology Nordic AB

        Comment


        • RandomJerk
          RandomJerk commented
          Editing a comment
          I agree. This is complete. I don't know if it's good, but it's complete.

          A is an axial feature, there's your primary, and locates two axes. B provides a rotational constraint around the axis formed by A, and the midplane of the two ends locks the third location.

          I don't know if the "Two coaxial weldnuts" are on opposite sides of the part, if so I'd be tempted to say cylinder between the two locknuts for both A and B, plane comprised of midpoints from the two ends for C. Pierce A and C and B and C, level to A, rotate between the two points, set the origin to the AC point.

          As for the perpendicularity of C to |A|B|, I'd create a midline of the two axial lines that are A and B and use that line to check C against.

        • KIRBSTER269
          KIRBSTER269 commented
          Editing a comment
          My apologizes, there isn't even a thru hole. according the second part of that print. So you have 2 nuts that need to align on one side of the plate, or one on each side of the plate?, but still what is the purpose of the position back to itself?

        • RandomJerk
          RandomJerk commented
          Editing a comment
          I'm guessing it's a bad way to try to call out perpendicularity.

      • #7
        SORRY I haven't responded I was stuck in a meeting. But if it helps the part is symmetrical, so it does have two other holes on the other end that match up to Datum A and Datum B

        Comment


        • KIRBSTER269
          KIRBSTER269 commented
          Editing a comment
          So could you be a little more clear on this PacMan23 just on the one side, how many nuts total?, 4? and is there a thru hole? If 4, one on each side?We understand the other side mirrors the A-A cutout.

      • #8
        KIRBSTER269 there are 4 total nuts(2 on each side), and they all have a thru hole (maybe not shown on the print).

        Comment


        • KIRBSTER269
          KIRBSTER269 commented
          Editing a comment
          Then that kinda goes along what I said earlier in my post. Let's us know how you approach it. With all the welding, I would restrain the part. unless otherwise instructed.

      • #9
        First thing I think we should tackle is to make sure to measure autocircles with that 1.75mm threaded pitch, any time you measure those weldnuts.


        Now start measuring: interim align to something simple. 3 hits on top flange, level z 2 hits along flange edge rotate y, 1 hit at end of part.
        Measure 2 circles on one side of part, name them cira1, cirb1 and outer plane holding the weldnuts (for datum c) name cplane1
        Measure 2 circles on other side of part, name them cira2, cirb2 and outer plane holding the weldnuts (for datum c) name cplane2
        Construct a line between cirA1 and cirA2, name it LineA,
        Construct another line between cirB1 and cirB2 name it LineB.
        Construct a perpendicular line between the two lines name it RotateLine.
        Construct a midplane between Cplane1 and Cplane2
        Level X to Line A
        Rotate Y to RotateLine about X
        Origin YZ to LineA
        Origin X to midplane.
        measure true positions and perpendicularity.
        Last edited by louisd; 12-06-2017, 04:44 PM.

        Comment


        • #10
          All 6 degrees are defined...terribly! But it's all locked up as it should be.
          PcDmis 2015.1 SP10 CAD++
          Global 7-10-7 DC800S

          Comment


          • #11
            I'm confused with the perpendicularity, can you attribute perpendicularity to a width like that? I seems kinda dumb, as it would be measuring perpendicularity to the midplane of the two planes.
            If planes were trapezoidal in shape, it wouldn't capture the variation. Parts would pass, yet it surely wouldn't work with whatever it's installing onto.

            I would think perpendicularity should be associated to each side, individually.
            Last edited by louisd; 12-06-2017, 04:45 PM.

            Comment


            • KIRBSTER269
              KIRBSTER269 commented
              Editing a comment
              This whole print is terrible. why are you just picking on that. That's why in one of my comments. I was telling him to check stuff that's not on the print, but I actually lacked to tell him this suggestion. I do agree with this also. PacMan23

            • PacMan23
              PacMan23 commented
              Editing a comment
              Yes I struggled with this all afternoon yesterday. Contemplating how much more of this I can handle, I'm definitely going insane, slowly but surely

          Related Topics

          Collapse

          Working...
          X