iSixSigma

Gage RR on Weld Testing

Six Sigma – iSixSigma Forums Old Forums General Gage RR on Weld Testing

Viewing 10 posts - 1 through 10 (of 10 total)
  • Author
    Posts
  • #28117

    wei
    Member

    Hi!
    I need to evaluate a destructive gage r&r study on a tensile pull tester, which is used to perform a pull test on the weld joint between a wire and a Stainless Steel marker band. Once the weld is destroyed, no more testing can be done on the parts anymore. Is anyone can kindly tell me how can I evaluate the r&r on this pull tester against the weld joint? Thanks!!!!
     

    0
    #69629

    Tommy
    Member

    MiniTab 13 accounts for destructive testing with the gauge r&r nested.

    0
    #69632

    Chaz
    Participant

    I am not that familiar with minitab. I have used Latin design. The quality of the analysis will not be the sam as from nondestructive MSA.
    For details type”destructive test” in the forum search box.
    Chaz Weyer

    0
    #69633

    gt
    Participant

    the idea is to work with an assumption. As each part are destructed, we cannot conduct a second test. So when comes time to choose the parts, pick consecutive parts for one readings. i:e: 3 parts will represent the same part  if you take 3 operators (that’s the assumption. we assume no change between these consecutives parts). Take 10 different (3 parts) sample and use minitab to study the R&R. In minitab, destructive testing is handled under nested gage R&R.
    rgds

    0
    #69643

    Ken K.
    Participant

    GT is correct.
    By manufacturing several (3) bonds without changing the equipment settings, there is an assumption made that these three bonds are nearly identical. Of course the variation measurement you obtain will actually be the combination of measurement variability PLUS part-to-part variability for very similar parts. There is really no way to get around this. For destructive tests this will always be confounded.
    The idea is that you are hoping that short-term part-to-part variability is much less than long term part-to-part variability. This may or may not be true.
    Another consideration is to find another way to excercise the gage and obtain a measurement without actually breaking parts. Try repeatedly hanging a weight from the device to see its repeatability – or something similar.

    0
    #69644

    Anonymous
    Participant

    Ken is right on the mark.Another simple test (can be used for any destructive test or in-line gage) is called Vendor P/T:(Vendor Std Dev * 6)/Process ToleranceNote this is not to be used as a substitute for GRR but can tell you if the equipment will meet Six Sigma requirements even before you purchase! Not well known, but used by the world’s largest semiconductor companies.Also don’t forget, the vendor should have homogenous material that they use for calibration purposes. This can be very helful in your GRR in order to minimize part-to-part variability.

    0
    #69645

    Same anonymous as above
    Member

    Useful references:Donald Wheeler, Evaluating The Measurement Process, 2nd Ed, page 15.Using Repeatability and Reproducibility Studies to Evaluate a Destructive Test Method
    Phillips, Aaron R.; Jeffries, Rella; Schneider, Jan; Frankoski, Stanley P.;
    Quality Engineering, Vol. 10, No. 2, DECEMBER 1997, pp. 283-290 Assessing Measurement Error in a Destructive Test
    Conklin, Joseph D.;
    47th Annual Quality Congress, May 1993, Boston MA, Vol. 0, No. 0, MAY 1993, pp. 400-405

    0
    #69646

    Anonymous
    Participant

    Note: Conklin has an earlier posting on this board.

    0
    #69654

    ofat guy
    Participant

    Here is some information obout a destructive test used in a bottle filling process:
    When performing a MSE destructive test one issue has to be resolved:
    How to simulate the within group repetitions. Two things are needed to simulate the repetitions
    1. Control chart of process parameter – this is to assure that the process is in control.  From the control chart you determine what is the expected total variation of the parameter.
    2. For a destructive MSE it is assumed that consecutive parts or samples have very little variation. For example three consecutive cans of the SAME valve would have a minimum variance, while samples of three consecutive valves would have a greater variance (valve to valve) or samples of the same valve two hours apart.
    For the MSE you need different parts and repetitions of the same part foreach operator.  To get different parts ; obtain samples from different valves at different times.  To get simulated repetitions obtain nine consecutive samples (three per operator) for each different part.
    Perform the MSE calculations.
    The MSE contribution is confounded with the operator to operator variance
    (reproducibility).  Compare the following ratio Total contribution from MSE – Gage R&R to observed process variation (from control chart r/d2)
    The MSE should be 20% or less than the total variation.

    0
    #69657

    Keith Boyle
    Participant

    The focus here should be on the tensile instrumentation. Do not use actual weld samples that have unknown variables. Instead use a  tensometer (stress/strain gage) of known value and accuractly calibrated. Sustitute this in place of weld samples and pull tests to a set values on the tensile tested. Record and analyse data from the tensometer. Repeat tests for different operators.
    Regards, Keith Boyle [email protected]
     

    0
Viewing 10 posts - 1 through 10 (of 10 total)

The forum ‘General’ is closed to new topics and replies.