Gage R&R for a Destructive Test - Weld Breakout Test

[RLBredeweg]

Gage R&R for destructive Test

What the test is is called is a "weld breakout test" they use a hydraulic press with a gage(Pressure meter) that reads in psi the part is fixtured and a ball on the end of the press comes down and pushes on the weld seam - Minimum specification for fracture is 3,000 Lbs. - No Max spec (infinity) or weld fracture. What I was considering is to verify the accuracy of the gage with a certified strain gage for correlation to the psi meter and than to review the last 30 days of measurement recordings (which I found out were being recorded as pass/fail)

I have to respond back for a CAR written by our Registrar ASAP Can any body give a suggestion?

My experience is with hardness measurements...may be similar, maybe not.

First, you should consider the gage R&R not in terms of the actual test as much as the unit being used for the test. Use (if available) a standard for your pressure machine. Use the same standard, across a variety of operators, and the classic GRR procedure to calculate operator and machine influence. (NOTE: This is NOT a true GRR, but it is a gage CAPABILIY.)

You can use the gage CAPABILITY in conjuntion with PROCESS CAPABILITY (your historical test values) to determine a true GRR.

This is too ornate to try to explain here...I suggest Concepts for R&R Studies by Larry Barrentine for a more thorough evaluation. Or send me private e-mail or give me a call...see my profile.

[Marc]

From Spaceman Spiff a while back:

-> The best method I found for MSA on destructive testing is
-> through using the ISOPlot method developed by Dr. Dorian
-> Shainin (see www.shainin.com for information).
->
-> Basically, I would take 2 consecutive samples as a group
-> and collect a total of 30 groups. Next I would use sample
-> 1 of each group as Trial 1 and sample 2 of each group as
-> Trial 2. Then I would plot Trial 1 versus Trial 2 and
-> determine the discremination ratio. It is very similar to
-> correlation, but the method of calculating discrimination
-> ratio is the key to qualify the measurement system.

The way I got around this (albeit for a Q1 audit) was to build a rig that simulated the thing being tested. So if we're applying force to a weld to destruction, build a rig that "gives" at a typical range of values, maybe by lifting a set weight through a lever system.

Cheers

Brian