Dg&t Wtf?

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

  • Dg&t Wtf?

    Got a part with a BS GD&T callout.

    Imagine if you will, a flat plate. It starts out basically as a rectangle with a small 'bump-out' at one corner. There is a hole in the plate. There is a 90 degree flange at ONE corner of the plate where the3 'bump-out' is. The flat of the plate is the A datum. The hole in the plate is the B datum. The edge of the plate is the C datum. They had a FCF pointing to the little 90 degree flange, but it is a 'double' callout out. 0.5 to ABC and 0.3 to C. WTF? The vector of this flange is -1,0,0. It can ONLY deviate from zero in the C datum direction. Why would they give it a double callout? It can not have any deviation in either the B or A datum vectoral direction.

    I guess I should call myself an engineer so I can be an idiot and make 3 times the money of the poor schlubs who will have to decipher my chicken scratches out.
    sigpic
    Originally posted by AndersI
    I've got one from September 2006 (bug ticket) which has finally been fixed in 2013.

  • #2
    Matt,

    Do not be so hard on yourself, even in the most P.Oed mood, I wouldn't call a friend an engineer. Especially on as smart as you. Sounds like the old, "I've seen this on a print before, and the parts worked well. Guess that is why, so I will put it on this one. We're going to the zoo, we're going to the zoo...."
    When in doubt, post code. A second set of eyes might see something you missed.
    sigpic

    Comment


    • #3
      Originally posted by John Kingston View Post
      Matt,

      Do not be so hard on yourself, even in the most P.Oed mood, I wouldn't call a friend an engineer. Especially on as smart as you. Sounds like the old, "I've seen this on a print before, and the parts worked well. Guess that is why, so I will put it on this one. We're going to the zoo, we're going to the zoo...."
      Heck, that ain't even the WORST of it! On an almost identical part, they call out the flat (a circle of 30mm diameter) the A datum as well as an edge point on the side of the plate. Now tell me, how in the world can you use 3 (or 4 or more) points with a vector of 0,0,1 and some with a vector of -1,0,0 all to make an A datum? I pointed all this out to the engineer here a month ago or more and told him to get with the customer to get an explaination of how this is even possible and I am still waiting for an answer and NOW I have parts coming in the door.
      sigpic
      Originally posted by AndersI
      I've got one from September 2006 (bug ticket) which has finally been fixed in 2013.

      Comment


      • #4
        Originally posted by Matthew D. Hoedeman View Post
        Got a part with a BS GD&T callout.

        Imagine if you will, a flat plate. It starts out basically as a rectangle with a small 'bump-out' at one corner. There is a hole in the plate. There is a 90 degree flange at ONE corner of the plate where the3 'bump-out' is. The flat of the plate is the A datum. The hole in the plate is the B datum. The edge of the plate is the C datum. They had a FCF pointing to the little 90 degree flange, but it is a 'double' callout out. 0.5 to ABC and 0.3 to C. WTF? The vector of this flange is -1,0,0. It can ONLY deviate from zero in the C datum direction. Why would they give it a double callout? It can not have any deviation in either the B or A datum vectoral direction.

        I guess I should call myself an engineer so I can be an idiot and make 3 times the money of the poor schlubs who will have to decipher my chicken scratches out.
        I do not see how it can be possible for a feature to be able to deviate in only one axis. It has to be able to deviate in at least two axis.

        Sounds like:
        A Plane (Primary) (3 DOF)
        B Cylinder Secondary (2 DOF)
        C Tertiary (Rotation)

        Further, assuming C is the X Axis Plane (a line along the Y Axis), it sounds like the feature in Question can deviate in the Y and the Z Axis at a minimum.

        The POS to ABC would be a straight Positional and the POS to C is Basically a Perp callout.

        Now, if C is the X Axis Plane the feature can deviate in all three Axis.
        Bill Jarrells
        A lie can travel half way around the world while the truth is putting on its shoes. - Mark Twain

        Comment


        • #5
          Originally posted by Wingman View Post
          I do not see how it can be possible for a feature to be able to deviate in only one axis. It has to be able to deviate in at least two axis.

          Sounds like:
          A Plane (Primary) (3 DOF)
          B Cylinder Secondary (2 DOF)
          C Tertiary (Rotation)

          Further, assuming C is the X Axis Plane (a line along the Y Axis), it sounds like the feature in Question can deviate in the Y and the Z Axis at a minimum.

          The POS to ABC would be a straight Positional and the POS to C is Basically a Perp callout.

          Now, if C is the X Axis Plane the feature can deviate in all three Axis.
          A datum is 0,0,1
          B datum is 0,1,0
          C datum is 1,0,0
          Feature in question is a flat surface with a vector of 1,0,0
          The ONLY direction that surface can deviate is in X, or the C datum direction. The edges of the surface can deviate in Y and Z but the surface itself can ONLY deviate in X, or the C datum axis.
          sigpic
          Originally posted by AndersI
          I've got one from September 2006 (bug ticket) which has finally been fixed in 2013.

          Comment


          • #6
            Originally posted by Matthew D. Hoedeman View Post
            A datum is 0,0,1
            B datum is 0,1,0
            C datum is 1,0,0
            Feature in question is a flat surface with a vector of 1,0,0
            The ONLY direction that surface can deviate is in X, or the C datum direction. The edges of the surface can deviate in Y and Z but the surface itself can ONLY deviate in X, or the C datum axis.

            So the feature in question is a surface that is not part of the C Datum? Correct?

            If so then the callout appears to say
            Align / Constrain to ABC and Check Feature Profile / Position
            Align ONLY to C and Check Profile / Position
            The could well be drastically different values based on the Perpendicularity of the C Datum to the other two datums.

            Regards,
            Bill
            Bill Jarrells
            A lie can travel half way around the world while the truth is putting on its shoes. - Mark Twain

            Comment


            • #7
              So why is this thread named Dg&t? I get the wtf part.
              <internet bumper sticker goes here>

              Comment


              • #8
                Originally posted by craiger_ny View Post
                So why is this thread named Dg&t? I get the wtf part.
                That's called mad-fat-finger-syndrome. And, it won't let you edit the 'title' of a post, so after you click SUBMIT it's too late to fix MFFS. Haven't you noticed that I have one hand that is faster than the other? Why else would there be so many 'teh' in my posts instead of 'the'? I suffer from FFS a lot!
                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 Wingman View Post
                  So the feature in question is a surface that is not part of the C Datum? Correct?

                  If so then the callout appears to say
                  Align / Constrain to ABC and Check Feature Profile / Position
                  Align ONLY to C and Check Profile / Position
                  The could well be drastically different values based on the Perpendicularity of the C Datum to the other two datums.

                  Regards,
                  Bill
                  Yes, the surface in question is NOT part of the C datum, but the C datum is only a single point on the edge of the part. This part has to be checked (for right now) using an iterative alignment because they have not yet gotten me a fixture. So, there isn't really a c-datum surface or edge, per se. 4 points on the surface around the hole, at a 15mm radius, the hole and a single point on the edge, this is to simulate how the fixture (so I have been told) will hold the part. Place the part on the flat, with the pin in the hole and rotate the part to a stop on the edge of the part then check. The vector for the point on the edge IS 1,0,0, the vector for the surface (a-datum) is 0,0,1 and the hole (b-datum) sets the XY origin, the C-datum (single point on edge) is teh clocking feature. Automotive sheet metal does this kind of datum scheme quite often so it's not like getting a cube, leveling to the top, rotating to the front, and setting XYZ origin to the corner.
                  sigpic
                  Originally posted by AndersI
                  I've got one from September 2006 (bug ticket) which has finally been fixed in 2013.

                  Comment


                  • #10
                    Originally posted by Matthew D. Hoedeman View Post
                    Yes, the surface in question is NOT part of the C datum, but the C datum is only a single point on the edge of the part. This part has to be checked (for right now) using an iterative alignment because they have not yet gotten me a fixture. So, there isn't really a c-datum surface or edge, per se. 4 points on the surface around the hole, at a 15mm radius, the hole and a single point on the edge, this is to simulate how the fixture (so I have been told) will hold the part. Place the part on the flat, with the pin in the hole and rotate the part to a stop on the edge of the part then check. The vector for the point on the edge IS 1,0,0, the vector for the surface (a-datum) is 0,0,1 and the hole (b-datum) sets the XY origin, the C-datum (single point on edge) is teh clocking feature. Automotive sheet metal does this kind of datum scheme quite often so it's not like getting a cube, leveling to the top, rotating to the front, and setting XYZ origin to the corner.
                    I am familiar with Automotive Sheet Metal and the fixturing schemes you are talking about. I used to work for a stamping plant and had to deal with it a lot there.
                    If there is no way to isolate the C Datum then the 2nd part of the composite tolerance is redundant / meaningless. Of course, they may expect you to hold the tightest of the two

                    Hope you have a CAD Model!!
                    Bill Jarrells
                    A lie can travel half way around the world while the truth is putting on its shoes. - Mark Twain

                    Comment


                    • #11
                      Originally posted by Wingman View Post
                      I am familiar with Automotive Sheet Metal and the fixturing schemes you are talking about. I used to work for a stamping plant and had to deal with it a lot there.
                      If there is no way to isolate the C Datum then the 2nd part of the composite tolerance is redundant / meaningless. Of course, they may expect you to hold the tightest of the two

                      Hope you have a CAD Model!!
                      Which was pretty much what I was planning on doing, what else can you do?

                      And YEP! I have cad models, I do on 99.9% of the stuff I work on and for those 1 or 2 in a blue moon where I don't have it, I can usually make it from the print, then use that. Not because I can't program from the print, but the others can 'see' the results on the tube if the part is there as well. And, you can then simply do a screen dump (with DIMINFO) and hand them that so tehy can make corrections, if needed.
                      sigpic
                      Originally posted by AndersI
                      I've got one from September 2006 (bug ticket) which has finally been fixed in 2013.

                      Comment


                      • #12
                        Check this one out.

                        Picture a Radiator Shroud with a center piece with three mounting holes. Primary Datum is defined as the three pads for the mount bolts (D). There is a hole through the center that the motor goes through. The hole is about 7 in diameter. OK, on the inner surface of the hole there are three '****' - 2mm Rad hemispheres. The tips of these spheres guide the motor and help hold it centered so the fan blade will not hit the barrel. Now, those three tips define a circle which is the secondary datum (E). One of the mount holes creates a tertiary datum (F). So far so good. But, then they call out the Secondary Datum (E) positional 0.3mm to the Primary Datum (D). WTF?? I can't even consider this as Perp because there are only three points defined. How can I check that nonsense? LOL
                        Bill Jarrells
                        A lie can travel half way around the world while the truth is putting on its shoes. - Mark Twain

                        Comment


                        • #13
                          Originally posted by Wingman View Post
                          Check this one out.

                          Picture a Radiator Shroud with a center piece with three mounting holes. Primary Datum is defined as the three pads for the mount bolts (D). There is a hole through the center that the motor goes through. The hole is about 7 in diameter. OK, on the inner surface of the hole there are three '****' - 2mm Rad hemispheres. The tips of these spheres guide the motor and help hold it centered so the fan blade will not hit the barrel. Now, those three tips define a circle which is the secondary datum (E). One of the mount holes creates a tertiary datum (F). So far so good. But, then they call out the Secondary Datum (E) positional 0.3mm to the Primary Datum (D). WTF?? I can't even consider this as Perp because there are only three points defined. How can I check that nonsense? LOL
                          Just report it, it will show ZERO so they will all think you are a god.

                          What about FO-MO-CO and their habit of placing SPC points ON TOP of the net pads? Every net pad on the fixture is called out and dimensioned on the print, and right next to it, AT THE SAME XYZ location, is an SPC point. Then they wonder HOW you can report a Ppk of 82,283 for a point on a sheet metal stamping. It's even greater when the net/spc are on non-mating surfaces, so your tolerance is +/-1.5 for the surface. Ya gotta love those!
                          sigpic
                          Originally posted by AndersI
                          I've got one from September 2006 (bug ticket) which has finally been fixed in 2013.

                          Comment


                          • #14
                            Originally posted by Matthew D. Hoedeman View Post
                            That's called mad-fat-finger-syndrome. And, it won't let you edit the 'title' of a post, so after you click SUBMIT it's too late to fix MFFS. Haven't you noticed that I have one hand that is faster than the other? Why else would there be so many 'teh' in my posts instead of 'the'? I suffer from FFS a lot!

                            I suffer from FFS as well. Remember, keyboards were designed by ENGINEERS with skinny, dainty, little girly fingers. Probably manicured nails too.

                            Consequently our fat fingers hit 2 maybe 3 keys at the same time.
                            Perry
                            B&S Mistral
                            3.207 Beta on XP

                            Older'n dirt

                            Comment


                            • #15
                              Some people cannot even spell GD&T.....(Ok Matt, are you exempt??)
                              kev
                              RFS Means Really Fussy Stuff

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

                              Comment

                              Related Topics

                              Collapse

                              Working...
                              X