hexagon logo
Register Sign in
  • Replies 9 replies
  • Subscribers 222 subscribers
  • Views 121 views
  • Users 0 members are here
Options
Related

Standard deviation

gitalife
Hello all,

Not being a statistician, what exactly is the standard deviation of each tip telling me.
Besides the formal definition of " a measure of how dispersed the data is in relation to the mean.",
if one of the tips has a standard deviation of .0002 (inches) and another .0005, how will that translate
to the accuracy of the real time measurements of those tips?
Is that for the deviation of the dynamic tip radius and does that differ from a qualification check which
seems to give a xyz and polar radius deviation of each tip relative to the sphere measurement location?

Thanks for any technical insight
Parents
  • Assuming that the calibration follows a statistical normal distribution, it would describe that 95% of hits are included into measured radius ±0.0004" for your first case, and ±0.001" for the second.
    You can measure the calibration sphere with the same number of hits than the calibration, same number of level, then look at the form or :
    ASSIGN/SCOPE=MAX(SQRT(DOT(SPH1.HIT[1..SPH1.NUMHITS].XYZ-SPH1.XYZ,SPH1.HIT[1..SPH1.NUMHITS].XYZ-SPH1.XYZ)))-MIN(SQRT(DOT(SPH1.HIT[1..SPH1.NUMHITS].XYZ-SPH1.XYZ,SPH1.HIT[1..SPH1.NUMHITS].XYZ-SPH1.XYZ)))
    Using another software could give you directly the scope of radii.
    I don't know what type of probe you use, and how many hits you take for a calibration, but I would say that 0.0005 " is a very bad calibration... (with a LSPX1 L50 D5, I usually get 0.0002 mm of stddev)
    The same value with 25 hits or 5 hits doesn't give the same evalution of the calibration...
Reply
  • Assuming that the calibration follows a statistical normal distribution, it would describe that 95% of hits are included into measured radius ±0.0004" for your first case, and ±0.001" for the second.
    You can measure the calibration sphere with the same number of hits than the calibration, same number of level, then look at the form or :
    ASSIGN/SCOPE=MAX(SQRT(DOT(SPH1.HIT[1..SPH1.NUMHITS].XYZ-SPH1.XYZ,SPH1.HIT[1..SPH1.NUMHITS].XYZ-SPH1.XYZ)))-MIN(SQRT(DOT(SPH1.HIT[1..SPH1.NUMHITS].XYZ-SPH1.XYZ,SPH1.HIT[1..SPH1.NUMHITS].XYZ-SPH1.XYZ)))
    Using another software could give you directly the scope of radii.
    I don't know what type of probe you use, and how many hits you take for a calibration, but I would say that 0.0005 " is a very bad calibration... (with a LSPX1 L50 D5, I usually get 0.0002 mm of stddev)
    The same value with 25 hits or 5 hits doesn't give the same evalution of the calibration...
Children
No Data
© 2024 Hexagon AB and/or its subsidiaries. Privacy Policy | Cloud Services Agreement