Gage R and R, Checking Fixture

Six Sigma – iSixSigma Forums Old Forums General Gage R and R, Checking Fixture

Viewing 5 posts - 1 through 5 (of 5 total)
  • Author
  • #49926


    I need help.  I need to perform a Gage RR study on a checking fixture.  The fixture measures several different dimensions for a vehicle carpet.  There are approximately 33 different inspection points on the fixture with (4) datum points.  I understand I have to conduct the study utilizing 30 parts, 3 operators, and 3 trials.  My question: what data point out of the 33 do I use to conduct the study?  Should I pick one point (gap and flush) or should I pick one of the holes with a width call out.  I’m lost on this one.  Please help.



    Check the following link for an explanation of R&R’s.  You can use 10 parts, 2 operators and 2 trials for the study, and all 33 points.  One thing that is not covered in the books is the part in the fixture in the unclamped state.  There is a maximum gap from part the datum that needs to be evaluated first, can you ask someone in your company what the standard is.  In a clamped state the part can be very good, unclamped it will not fit the assembly.
    Check out the following links.  AIAG has a book that everyone used as the standard for how to do the study.  Search the web for Excel based worksheets to help you do the analysis on Minitabs and other software packages can do it.

    Industry standard on R&R.
    Book Title
    Measurement Systems Analysis Product Code: MSA-3
    Hope this helps.



    Hi Tom,
    When I conduct a R&R study on a checking fixture which has a lot of control points, I use a 2 operators, 5 pieces and I choose several control points ramdonly but with the suitable sample size confidence level.
    I hope this can be useful. 



    When checking a fixture, you will need to check every dim that is applicable to the fixture, whole point of the R&R. Use 10 parts and 3 operators. Important to have 3 operators, due to possilbe issue of checking same parts over and over, this will trend if its an issue. Also check them in the same sequence each time. If you do it this way, you will have results that are meaningful.



    Be careful which points you choose here. Measuring a hole and a slot is really difficult as there are at least 4 methods you can use, each will give you a different answer. So I would go for fixed points.
    I would look first at the types of measures you need to make, IE locations far apart, on/off orthagonal plane and relevance to a CC/SC dimension, etc. Once I have the critical measurement points I would run the MSA. Then use the data to understand if the MSA need adjustment and this will have be done my Operational Definitions which means that each point had to be measured in a unique manner. I may even study the carpets to see where the shape/form is not held in the mould and ensure these are NOT in my MSA as these will mess you up.
    Also looking at the compliance in the carpet, this may throw the gauge, so I would look to simulate it in hard points. If you can measure these then you have a chance of measuring the carpet.
    I did something similar on measuring an engine bay and ended up have 5 Operational Definations in an MSA to get the gauge to work.

Viewing 5 posts - 1 through 5 (of 5 total)

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