Measuring a "calibration" process of a machine

[SJAQM]

Morning all,

First time posting so please forgive me if I am not quite up to speed with the forum lingo.

I have been tasked with demonstrating the control/capability of the calibration process of a machine. A test is run daily at the start of each shift, to verify that the machine is cutting as required. The template that is created from the test is then offered up to a Master template and if everything is OK the machine is set away cutting for the day.

There are no tolerances applied when offering up of the template to the master. It is simply OK (Pass) or not OK (Fail).

I have suggested presenting the data using a chart in Excel but the opinion from other areas is that this can be achieved using SPC charts. I use Minitab for monitoring the capability of another process and ideally would like to use this.
The method I used in Excel seemed quite crude but I feel it worked I have attached a rough cut version of sample data.

I would appreciate any help or advise on ways to present this better etc.

Thanks in advance,

 

[AndyN]

The study of a machine process is a capability study, not a "calibration". If you use that specific term it will lead to all manner of complications...

Taking one item off a machining process is, relatively, meaningless. It's much more effective - and builds confidence in results overall - to run a study of 50 items and measure those to see what inherent variation comes from the process...

 

[SJAQM]

Thanks for that. I might give that a try.

 

[Sean Kelley]

This depends a lot on the customer and end use of the product. If you are creating a yard ornament that only needs to visually look similar then what you are doing is likely good enough. However if you are making brake pads and that plate must fit into a caliper then dimensions are more critical. The key is to determine how critical is the feature you are measuring and then determine what equipment is needed to meet the measurement.