Hi all,

Good to find a source of information from people in the same "boat".
I have been asked to carry out an MSA study on a tensile tester as it is identified in one of our control plans.

I know the QS9000 MSA handbook makes reference to this type of study but it does't make much sense to me. Can anyone put the steps of the study into plain english.

Thanks

Hello Mattye and welcome to the cove. It's difficult to do an MSA study on a destrutive test such as tensile testing. For my 2 cents, you're better off participating in an interlaboratory study to compare your testing results versus that of other laboratories using standard materials. I'd also check the method you're basing your test on, and see if it does not include an accuracy statement. Here in the US, test methods are normally based on ASTM standards and the test method for tensile testing of metallic materials is E8, which includes statements on precision and bias. Within that method, they recommend MSA studies only on micrometers and calipers used to measure prepared tensile specimens, which is what I did as manager of a laboratory accreited to ISO Guide 25 for mechanical testing, including tensile testing. I also took part in interlaboratory studies that compared the results of my laboratory versus a number of other laboratories for both tensile testing (which included sample preparation) and for Rockwell hardness testing (separate interlaboratory programs).

It's been about 6 years since I reviewed European testing standards , so I do not remember requirements in them, nor if they include precision and accuracy statements as do the ASTM standards. Possibly one of the other members may be able to help in that area.

Good luck.

Can't you just have the tensile machine and load cells and the calipers/extensometers calibrated by a UKAS accredited calibration lab and push on? Let the calibration lab do the work, they will even provide the measurement uncertainty.

Just my thoughts. Hope it helps.

Hershal

Hi all,

Good to find a source of information from people in the same "boat".
I have been asked to carry out an MSA study on a tensile tester as it is identified in one of our control plans.

I know the QS9000 MSA handbook makes reference to this type of study but it does't make much sense to me. Can anyone put the steps of the study into plain english.

Thanks

In case of distructive testing, you need to take samples as uniform as possible. For example, if you want to consider 3 operators and 2 trials you need 6 samples to be tested. Take 6 samples for tesile testing from the same batch (produced under identical conditions). There will be some with-in part variation and is acceptable.
Similarly, collect 5 or 10 parts. In this case one part means 6 test samples collected under same conditions. Then rest of the Gage R&R method can be used for the analysis.
I have used this approach for Hardness testing also. It works!