Measuring Circles on CMM - Standards? Good practices?

  • Filter
  • Time
  • Show
Clear All
new posts

  • Measuring Circles on CMM - Standards? Good practices?

    Does anyone have any standards or references that you use when measuring diameters as far as how many points? I'm looking for something that I can set as a standard for diameter ranges. For example, 0-1" - 8pts, 1-5" 20pts, etc.
    I used to be high on life but I built up a tolerance.

    Brown & Sharpe Global Advantage
    PCDMIS CAD++ v2011mr2


  • #2
    We have no standard, generally don't use the CMM for measuring diameters.
    We don't use true position anymore either, we calculate the TP manually.
    I don't think you can standardize number of hits unless all your holes are standardized as far as roundness.
    If I had to come up with a method to determine if I have enough hits, it would be when additional hits do not change the measured diameter (wonder if this could be made into a loop).
    Last edited by RussL; 02-01-2007, 10:41 AM.


    • #3

      I'm using version 3.7mr3. How do you go about calculating "tp" manually?

      Are you using some form of variables etc., if so please explain?

      What about if there is a modifier such as this; |1.0|A|Bm|Cm| ??

      If you have examples please email me at; [email protected]


      • #4
        I used to trig it out - is time consuming.
        Now, I just plot the measured value back in the customer model,
        and measure the distance perpendicular to the feature center-line.
        we then allow for MMC and/or LMC.


        • #5
          Your process would likely have the greatest influence on how many hits to use, after that your tolerance. If your process creates diameters with certain form conditions then you'll likely have to be strategic about number of hits and the location of hits. Tolerance will dictate how many and where for obvious reasons. I don't think there is a generic protocol for number of hits or that it would be safe to edevor adopting one.
          <internet bumper sticker goes here>


          • #6
            I have a similiar "issue". We have a customer that recently rejected some parts for the angle of a cone being out of tolerance. I created a program, that used 18 hits per level to create the cone in DCC. Pretty straight forward, not too difficult. Did my alignment, etc, etc. All the stuff that needs to be done to measure a feature accuratly and correctly. Well we asked them to let us know how they were checking this part in their incoming inspection, so we could check it the same way as they were. Let me say that this customer is located "south of the border" and were using GeoPak 3. Anyway, I got this very nice powerpoint presentaion of how they were checking this angle. Turns out they were using 6 hits total (3 on the bottom, 3 on the top) in manual mode. Now I do not claim to be a CMM expert, or guru, or anything remotely close to that, but I do know that 6 hits for a 5inch cone is no where near being accurate as it should be. So, is the customer right?? This kind of thing just makes me mad as heck. Come on, anybody with any true CMM experience would know better.

            Sorry, maybe I'm just venting.........


            • #7
              They probably weren't vectoring well at that.
              <internet bumper sticker goes here>


              Related Topics