Dimensioning slots NO FCF equation only, please advice

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

  • Dimensioning slots NO FCF equation only, please advice

    Okay, so I need to define these slots, here is the notes. Anyone put something together for something like this before? haha.
    I tried reaching out to engineering for Engineering intent but he doesn't have an effing clue. Legacy stuff and I just wanna make sure I have it done correct.
    disregard note 13, its about deburring only.


    pdf_1.JPGpdf_2.JPG

    -riley Lawlor
    Southern California
    Aerospace CNC Machinery

  • #2
    Post a picture of the slot/s, are they "straight" or "jellybean-shaped"?
    PC-DMIS CAD++ 2o19 R1 SP11

    Comment


    • rtlawlor
      rtlawlor commented
      Editing a comment
      thank you!! Post pic below

  • #3
    slots.jpg
    -riley Lawlor
    Southern California
    Aerospace CNC Machinery

    Comment


    • #4
      PLEASE HELP!! haha
      -riley Lawlor
      Southern California
      Aerospace CNC Machinery

      Comment


      • #5
        Well, this isn't GD&T - that's for sure. If you want to pursue what is written as tolerance on drawing, google "slope-intercept form" or "y=mx+b". Personally, I'd demand a GD&T tolerance to check these slots for - not some kind of math-plot-algorithm.

        I don't know about you, but if it isn't GD&T, it ain't my job...

        It is possible that JEFMAN could have something up his sleeve, as the math wiz he is. ​​​​​​​
        PC-DMIS CAD++ 2o19 R1 SP11

        Comment


        • #6
          boundary, that is what it sounds like they want, but that is just a guess.
          sigpic
          Originally posted by AndersI
          I've got one from September 2006 (bug ticket) which has finally been fixed in 2013.

          Comment


          • #7
            I agree with Matthew D. Hoedeman , reads like it was written by a six sigma statistician guru.
            request a GD&T translation .

            Comment


            • #8
              It looks to me like you will need to probe opposing points on the interior of the slot widths, export the center point data to something like Excel, and perform a basic linear regression on the point data for the slots. This regression line will then be used to test conformance for each slot to the desired specification. For example, the slope (m) of the best fit line must be within the parameters specified in note #9. The standard deviation of your point data relative to your regression line must be within the limits from note #9. The end of note #9 also lists limits on the y-intercept (b) that must be met. Now, for each slot, the actual positional tolerance is a zone +/- .0006 about the regression line. Each slot's individual point data would need to be analyzed to determine conformance to the six-sigma zone about the regression line (a maximum of 5 slots can be within a +/- .0010 zone).

              As vpt.se says, however, this is really outside of the scope of GD&T/PC-DMIS. I would request modern GD&T controls.
              Last edited by JacobCheverie; 08-19-2020, 08:23 AM.

              Comment


              • JacobCheverie
                JacobCheverie commented
                Editing a comment
                If I am not mistaken, the following code will perform a basic linear regression fit on features with point data.

                Code:
                ASSIGN/V1=1
                ASSIGN/SXY=0
                ASSIGN/SX=0
                ASSIGN/SY=0
                ASSIGN/SX2=0
                ASSIGN/NUMHITS=0
                ASSIGN/NUM1=SLOT1.NUMHITS
                ASSIGN/NUMHITS=NUMHITS + NUM1
                ASSIGN/NUM2=SLOT2.NUMHITS
                ASSIGN/NUMHITS=NUMHITS + NUM2
                
                etc....
                
                WHILE/V1 <= NUMHITS
                
                   IF/V1 <= NUM1
                      ASSIGN/X=SLOT1.HIT[V1].X
                      ASSIGN/Y=SLOT1.HIT[V1].Y
                   END_IF/
                   ELSE_IF/V1 <= NUM1 + NUM2
                      ASSIGN/X=SLOT2.HIT[V1 - NUM1].X
                      ASSIGN/Y=SLOT2.HIT[V1 - NUM1].Y
                   END_ELSEIF/
                
                   etc...
                
                   ASSIGN/SXY=SXY+X*Y
                   ASSIGN/SX=SX+X
                   ASSIGN/SY=SY+Y
                   ASSIGN/SX2=SX2+X^2
                   ASSIGN/V1=V1+1
                
                END_WHILE/
                
                ASSIGN/M=(NUMHITS*SXY-SX*SY)/(NUMHITS*SX2-(SX)^2)
                ASSIGN/BVAR=(SY-M*SX)/NUMHITS
                
                COMMENT/REPT,
                      "SLOPE m = " + M
                      "Y-INTERCEPT b = " + BVAR

                This will work to take all the X, Y point data from each slot and compute the regression line that passes through all slots.
                Last edited by JacobCheverie; 08-19-2020, 10:02 AM.

            • #9
              I will try that. Yeah I had a feeling I might need to just export the x.dev to a txt/xlsx file, and have a script written that would take that data and put in the equation in excel and solve it then autosave it. I will try what Jacob did tho
              -riley Lawlor
              Southern California
              Aerospace CNC Machinery

              Comment


            • #10
              I ended up posting out the x-dev, y dev, & slot length to a csv file a1-196, b1-196, c1-196.
              I still need to find a way to just have a script that opens up my spreadsheet and auto dumps all those variables in a,b&c, and then saves it to the variable path I have for path file destination. But I havnt figured that out yet. Left room for improvement. So as of now everyday when I come in I just open all the parts the ran the night before to check the linear regression in my excel doc I made. its a pain in the butt, takes like 10 mins to do 5 parts. but it is what it is. I have a couple scripts I use for other things that people wrote me. im going to try and look at that they did and replicate it for what I need to do I guess.

              THANK YOU EVERYONE!!!!
              -riley Lawlor
              Southern California
              Aerospace CNC Machinery

              Comment

              Related Topics

              Collapse

              Working...
              X