iSixSigma

Robert Tipton

Forum Replies Created

Forum Replies Created

Viewing 6 posts - 1 through 6 (of 6 total)
  • Author
    Posts
  • #252537

    Robert Tipton
    Participant

    Hi

    +/- 0.005 allows for 0.005 in the X Direction, as well as +/-0.005 in the Y Direction.

    This Equates to a Hypotenuse, at a 45 Degree Angle, of 0.007  (Radius) Times 2 = .014 True Position RFS.

    Not Considering the Tolerance for the MMC, the 0.002 MMC, would be a severe and dramatic reduction of the total tolerance.

    0.002 True Position is Equal to a Radius pf 0.001, which Equates to +/- 0.0007 in the X and Y Direction.

    Again, Huge Reduction in the Original Tolerance

    If you need more detail, Please contact me via e-mail

    RDT

     

     

     

     

     

    0
    #252202

    Robert Tipton
    Participant

    Hi –

    The General Rule of Thumb is +/- 25% of the Sample Size.

    Example: Average Sample Size of 100 +/- 25%  =  +/-25,  it is OK to use the same Control Limits.

    If the Sample Size is <=74, or >=126, then use Different Control Limits.

    These different control limits are sometimes referred to as a Manhattan Skyline.

     

    0
    #251108

    Robert Tipton
    Participant

    Sigma is the Standard Deviation, not the formula you indicated

    0
    #251107

    Robert Tipton
    Participant

    Hello –

    If you have a negative CPk, it means you have at least 50% Failures, assuming the formula is correct.

    The lower the negative CPk, the more Percent failures.

    As an Example;  Negative 1 CPk would be about 99.7% Failures, to the Specification

    Then look at the CP Value.  If CP is positive, you probably only need to adjust the mean (X-Bar), to center the process.

    It is possible to have a negative CPk, with a positive sigma (Std. Dev.) as the Std. Dev. will always be a positive number.

    The CPk value is the Real Calculated Value

    The CP Value is a Theoretical Value, Assuming the Process is perfectly centered around the Nominal Dimension.

    Hope this helps

     

     

     

     

    0
    #247660

    Robert Tipton
    Participant

    I have done hundreds of leak testers, with no problem

    assure that you have a maximum of 10% gage resolution.

    use normal Gage R & R, with Normal Bell Curves.

    Use SPC with Normal or Pearson Curve

     

    0
    #247130

    Robert Tipton
    Participant

    Hello –

    I would like to provide some additional insight.

    1.) Assure that the Gage is Acceptable for the measurements.

    1.1) Check Increments and Resolution, Gage Resolution should be less than 10% Resolution, I recommend 5%.

    Example: +/- .010 = Total Tolerance of .020.   Increments to be less than .001 = 5% Resolution.

    1.2) Gage Accuracy, Check a Known Value at Low End and High End of Specification.  This assures the Gage is Accurate at both values.

    1.3) Gage R. & R. , ideally less than 10% of the Part Tolerance, or less than 20%, never more than 30%.

    Variable Data – Long Method 10 Pieces, 3 Operators, measured 3 times, is the preferred method.

    2.) Short Term Capability Studies, begin with 15-30 pieces, depending upon cost

    2.1) Optimize process and complete a new Capability Study, Comparing the PPk Values.

    2.2) do Design of Experiments to Improve Process, then New Capability Studies.

    2.3) After diligently trying to improve the process, than ask for relaxed tolerance.

    Hope this help

    RDT

     

    0
Viewing 6 posts - 1 through 6 (of 6 total)