Note 9 part 3 states “Riveted Joint Quality Control Per QCI 9310-398. Riveted Joint to withstand a torque of 9Nm when rotated about the axis of the item 2 pin.”

This test is accomplished by first welding a nut of appropriate size to the external crank pin. A clear mark should be applied to the crank pin and the crank arm so that twisting can be visually identified. The external crank arm is held in a vice and a slow steady increase in torque is applied to the welded nut with a torque wrench until twisting has visibly occurred at the mark. The torque at which twisting occurred at should be noted. A torque wrench with a “tattle tale” dial is recommended as an accurate way to record the maximum torque required.



Because this is a destructive test (ie. the second time you apply a torque to the same part the reading will be lower) parts cannot be reused for this GR&R test. Following the six sigma handbook as a guide for GR&R with a destructive test, it is recommended to use parts that have different distributions. These don’t necessarily all have to be cranks, but could be bolts twisted to failure. Below is a suggestion of how to accomplish this:
• Use 5 parts, 3 operators and 2 trials
6 pieces each of the following parts will be required:
 Part 1: riveted crank pin
 o Part 2: welded crank pin
 o Part 3: 4 mm bolt and nut (torque to failure)
o Part 4: 6 mm bolt and nut (torque to failure)
o Part 5: 8 mm bolt and nut (torque to failure)
• The bolt sizes above are just recommendations, but they should be sized appropriately so that failure torque is within the range of the torque wrench you are using. The same torque wrench should be used for all tests.
• Once the matrix of measurements below is completed the GR&R of your process can be calculated.


Part 1 Part 1 Part 2 Part 2 Part 3 Part 3 Part 4 Part 4 Part 5 Part 5
Operator Trial 1 Trial 2 Trial 1 Trial 2 Trial 1 Trial 2 Trial 1 Trial 2 Trial 1 Trial 2
1
2
3

可以采用NON REPLICABLE GR&R, Minitab里面用GR&R Nested来做．
个人建议应该先做一下CpK或者PpK, 如果结果OK再做GR&R, 否则没有意义．