Using profile of a surface

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

  • Using profile of a surface

    I have a part that has a 2.5 radius call out on a really small area, it is almost imposible to get an acurate reading,
    what i did is measure a profile instead, using the Cad model provided by the costumer;
    doin this i am getting good results...

    my question is: in theory, is this a valid option?

  • #2
    Originally posted by hitalov View Post
    I have a part that has a 2.5 radius call out on a really small area, it is almost imposible to get an acurate reading,
    what i did is measure a profile instead, using the Cad model provided by the costumer;
    doin this i am getting good results...

    my question is: in theory, is this a valid option?
    The measurement of the radius as a profile is valid and probably the best way to do it but would have to be approved by your customer. Its not correct just to make this decision yourself. Plenty of threads here that explain that measuring Radii with <180 degrees of arc will give inaccurate results.
    3.7mr3CAD++ / 2011mr1CAD++/2012mr1CAD++/QUINDOS7

    Comment


    • #3
      I use this method all the time. Although I would have to disagree with the ( sorry Liambo ) earlier post. Why would you need permission? You are verifying a feature to print and you should stand by your measurement no matter what method you are using. You could use a rad gage that would tell you "it's close enough", but you need permission to use profile on your cmm ? Unless you have an agreement with the customer on how it will be actually verified, I say its up to you.

      Comment


      • #4
        Originally posted by Schlag View Post
        I use this method all the time. Although I would have to disagree with the ( sorry Liambo ) earlier post. Why would you need permission? You are verifying a feature to print and you should stand by your measurement no matter what method you are using. You could use a rad gage that would tell you "it's close enough", but you need permission to use profile on your cmm ? Unless you have an agreement with the customer on how it will be actually verified, I say its up to you.
        A radius with a tolerance is not the same as a profile with a basic radius. That's why.

        A profile allows for greater variation of the actual "measured radius" while maintaining location and form within boundaries. A directly tolerance radius allows the center to move, and the size to change, making for a much different shaped theoretical tolerance boundary (there is actually not a boundary, per se, in a directly toleranced radius.)

        They are NOT directly interchangeable, and should not be done so without express permission. They both have their uses, where one is better suited for a particular application than another.
        "This is my word... and as such is beyond contestation."

        Comment


        • #5
          Originally posted by Schlag View Post
          I use this method all the time. Although I would have to disagree with the ( sorry Liambo ) earlier post. Why would you need permission? You are verifying a feature to print and you should stand by your measurement no matter what method you are using. You could use a rad gage that would tell you "it's close enough", but you need permission to use profile on your cmm ? Unless you have an agreement with the customer on how it will be actually verified, I say its up to you.
          The OP does not have design authority over the print. A radius measurement is not a profile measurement, permission must be granted to deviate from the print callout. Just saying.
          3.7mr3CAD++ / 2011mr1CAD++/2012mr1CAD++/QUINDOS7

          Comment


          • #6
            Ok. Its been a long week and my brain does hurt already. I will take my lumps on this for discussions sake. So if you have a RAD callout ( no location here , just for discussion on this example ) it would be acceptable metrology practice to maybe use a rad gage ( or pin ) or comparator , or various other methods, but using form only profile callout would violate the print and would need permission ? 1st post didn't really clarify exact the callout ?

            Comment


            • #7
              2.5mm radius. what tolerance?

              let's ASSume a +/-0.25mm tolerance for that radius.

              What tolerance would you use for profile? Can't use +/-0.25

              If this radius is 1/4 of a circle (90 degrees) and you use +/-0.25 for the profile tolerance, and at the 'left' side it is +0.2, at the 'right' side it is +0.2 and in the middle it is -0.2, you would have a radius of 1.746, which fails. If it was -0.2, +0.2, -0.2 (opposite of first example) the radius would be 5.069, which also fails.

              A radius is a radius, and that is how it must be checked, UNLESS the 'owner' of the part/print decides that "Well, OK, the radius isn't the critical thing, just if it is within a +/-0.25 band of nominal" which would be profile.
              sigpic
              Originally posted by AndersI
              I've got one from September 2006 (bug ticket) which has finally been fixed in 2013.

              Comment


              • #8
                Another way:
                If you can determine where the center of this radius should be, you could set up an alignment with that center as your alignment origin.
                Then measure polar points using the radius size as nominal and apply whatever tolerance is given for the radius.
                I have done this in the past and felt no need to confirm applicability with the customer.
                Lately, it occurs to me
                What a long, strange trip it's been.

                2017 R1 (Offline programming)

                Comment


                • #9
                  Originally posted by hitalov View Post
                  I have a part that has a 2.5 radius call out on a really small area, it is almost imposible to get an acurate reading,
                  what i did is measure a profile instead, using the Cad model provided by the costumer;
                  doin this i am getting good results...

                  my question is: in theory, is this a valid option?
                  Depends on what radius you are talking about.
                  Tangent radius tolerance definition has changed since 1994, R symbol is now meant CR, look it up.
                  Using profile for this type of radius a big no no, best fit method isn’t right either nor taking hits from theoretical centerline.

                  Comment


                  • #10
                    I agree, this is the exact problem I have encountered last week. Our drawing have a SR callout with. A Location of the Radius. I measured the sphere and gave them the engineering the results. They came back saying that they want a profile, which is not on the drw. I gave them the Profile form+loc, they were happy. I tried to explain to them that if they want a Profile they need to add this to the drw. They refused.
                    Originally posted by VinniUSMC View Post
                    A radius with a tolerance is not the same as a profile with a basic radius. That's why.

                    A profile allows for greater variation of the actual "measured radius" while maintaining location and form within boundaries. A directly tolerance radius allows the center to move, and the size to change, making for a much different shaped theoretical tolerance boundary (there is actually not a boundary, per se, in a directly toleranced radius.)

                    They are NOT directly interchangeable, and should not be done so without express permission. They both have their uses, where one is better suited for a particular application than another.
                    B&S One
                    PC-DMIS CAD v2014

                    Romer Infinity

                    Comment

                    Related Topics

                    Collapse

                    Working...
                    X