Does this profile pass or not??

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

  • Does this profile pass or not??

    Take a look at PDF attachment....

    I set it to show "Only Out of Tolerance"....I have a green bar. I do realize that my tolerance is +/- .010 and the deviation is above that. Is that why it shows up OOT? Even though the max and min are less than .010???
    Last edited by jkb_mcg; 10-06-2009, 10:06 AM.
    ~Josh~ Giles Rescue "Ut ceteri vivant"

  • #2
    The measured band is 0.012 but the deviation is 0.008 and 0.007 in one direction and 0.004 and 0.004 in the other. The Demon reports it goofy like this. The profile tolerance is 0.020 and the measured profile is 0.016 and 0.011. The parts are good. IMO
    sigpic GDTPS - 0584

    Comment


    • #3
      Originally posted by jkb_mcg View Post
      Take a look at PDF attachment....

      I set it to show "Only Out of Tolerance"....I have a green bar. I do realize that my tolerance is +/- .010 and the deviation is above that. Is that why it shows up OOT? Even though the max and min are less than .010???
      ...you are reporting "Form and Location" the software is comparing the measured values back to the actual datum locations....make sure your datum nominals are correct and if they are machined datums make sure they are as close to nominal as possible...w/ that being said, I don't use the profile dimensions a great deal and I may be overlooking/missing something...or, maybe, I might be just wrong.... the minus tolerance is applied to the "low" points and the plus tolerance is applied to your "high" points...

      Comment


      • #4
        Pcdumbass, depending on the version you are using, will either report the profile, ONE of two ways (FORM & LOCATION)

        #1) Max - Min deviation BUT, if it is standard +/-, it will tell you if either the MIN or MAX is out of tolerance.

        #2) Largest deviation is DOUBLED for the profile result, THUS acting like TP, the restul is the MINIMUM tolerance needed for the feature to be in tolerance.

        Easiest way around this? Dimension individual points. If it is profile ABC (straight bi-lateral) of 0.020", then you have an individual point tolerance of +/-0.010.
        sigpic
        Originally posted by AndersI
        I've got one from September 2006 (bug ticket) which has finally been fixed in 2013.

        Comment


        • #5
          Try to dimension it with T-value of each hit. If max of T-val < 0.01 and min >-0.01, then part is good.

          Comment


          • #6
            graph the scan and ask for the min and max flags. This should be a good visual for whats going on.
            sigpic
            3.7 MR1
            4.2 MR1
            6 yrs with Metrolog 3 with PCDMIS

            Comment


            • #7
              Thanks for all your replies. I thought it was good. And figured it was the demon doing this crap.
              ~Josh~ Giles Rescue "Ut ceteri vivant"

              Comment


              • #8
                To get a report of method #2 as noted by Matt above:

                "#2) Largest deviation is DOUBLED for the profile result, THUS acting like TP, the restul is the MINIMUM tolerance needed for the feature to be in tolerance."

                Add this code after each Profile dimension:

                Code:
                ASSIGN/V1 = ABS(PROF1.MAX)
                            ASSIGN/V2 = ABS(PROF1.MIN)
                            ASSIGN/V3 = (V1*2)
                            IF/V1<V2
                            ASSIGN/V3 = (V2*2)
                            END_IF/
                            COMMENT/REPT,+++++++++++++++
                            COMMENT/REPT,"Extended PROF =   " + V3
                            COMMENT/REPT,+++++++++++++++
                What this does is double the greatest absolute deviation found in the group of points included in the profile.

                You will need to manually type in the the name of the profile dimensin for each. IE; PROFX will need to change for each one.

                An example of what this looks like in a report using legacy dimensions:



                Note: rounding may affect the extended profile value by a tenth.
                This will not be needed if you are using 4.x.
                With that you can choose the reporting method from within the Settings Editor.
                Last edited by John Riggins; 10-22-2009, 03:34 PM.
                Lately, it occurs to me
                What a long, strange trip it's been.

                2017 R1 (Offline programming)

                Comment


                • #9
                  Let me bring a section of my drawing into this. Perhaps I need to do it a different way. Notice the profile callout on the bottom left. I assume -D- is the pattern of the 4 holes. So what I do is, I check all 4 holes to A,B,C and report the T/P. Then I do a BF Alignments to all 4 holes. And report the holes again. After that, I keep the same alignment and measure various edge points around the top and bottom sections. Then I create Feature Sets. Then, do profile "Form and Location" because its referencing more than one Datum. Does this sound like the right approach?


                  http://www.pcdmisforum.com/attachmen...1&d=1245258083
                  Last edited by jkb_mcg; 10-06-2009, 10:06 AM.
                  ~Josh~ Giles Rescue "Ut ceteri vivant"

                  Comment


                  • #10
                    Yes.
                    That is what I would do.
                    As long as you are still "leveled" to -A-.
                    Lately, it occurs to me
                    What a long, strange trip it's been.

                    2017 R1 (Offline programming)

                    Comment


                    • #11
                      Yep. Still leveled to it.

                      I guess I am...When I went it to do the Best Fit I just clicked Best Fit and added the 4 circles. Then hit Ok....And Ok. Do I need to level and zero on Datum A again?
                      ~Josh~ Giles Rescue "Ut ceteri vivant"

                      Comment


                      • #12
                        If you did a 2D BF, with rotate and translate, you should be good to go!
                        Lately, it occurs to me
                        What a long, strange trip it's been.

                        2017 R1 (Offline programming)

                        Comment


                        • #13
                          Just checked....I did. Thanks John.

                          I'm going to mess around with the other code you posted earlier.
                          ~Josh~ Giles Rescue "Ut ceteri vivant"

                          Comment


                          • #14
                            I assume this is correct?

                            Code:
                            DIM PROF1= PROFILE OF LINE OF SET SCN1    FORMANDLOCATION  UNITS=IN ,$
                            GRAPH=OFF  TEXT=ON  MULT=20.00  OUTPUT=BOTH
                            AX       MEAS       +TOL       -TOL        MAX        MIN        DEV     OUTTOL
                            M       0.012      0.010      0.010      0.008     -0.004      0.012      0.000 --#|---#
                                        ASSIGN/V2 = ABS(PROF1.MAX)
                                        ASSIGN/V3 = ABS(PROF1.MIN)
                                        ASSIGN/V4 = (V2*2)
                                        IF/V2<V3
                                        ASSIGN/V4 = (V3*2)
                                        END_IF/
                                        COMMENT/REPT,**************************
                                        COMMENT/REPT,"Extended PROF =  " + V3
                                        COMMENT/REPT,**************************

                            http://www.pcdmisforum.com/attachmen...1&d=1245262104
                            Last edited by jkb_mcg; 10-06-2009, 10:06 AM.
                            ~Josh~ Giles Rescue "Ut ceteri vivant"

                            Comment


                            • #15
                              You have something wrong.
                              The Extended PROF should = .016
                              Lately, it occurs to me
                              What a long, strange trip it's been.

                              2017 R1 (Offline programming)

                              Comment

                              Related Topics

                              Collapse

                              Working...
                              X