Instruments that have to be documented in Control Plans

[omolina]

Hi. I need some advice. I'm confused about the instruments we have to document in a control plan. I suppose that at least they are the ones we use to measure some characteristic of our product, but what about the ones that we use to control our process. I will promptly accept to calibrate both groups, but doing R & R studies on them is another issue. The problem shows up because of clause number 4.11.4

This requirement shall apply to measurement systems referenced in the Control Plan.

If we have to document all the instruments we use to control our process R & R studies will be endless. What are the arguments for not doing that ?

Thanks...

 

[Al Dyer]

There is no argument for not doing an R&R study. How could you possibly ensure that your product meets specifications unless you have accurate gaging. And not only R&R, but Bias, Linearity, and stability?>???

If your looking for a way out I'm sure you can find it, but isn't the reason we are in this business to make sure things are done right!

Sounds to me like your company is trying to circumvent some basic principles!

With Respect,

 

[omolina]

Al:
Thank you for your comments, but I think I was missunderstood. Lets analyze the following examples:

1) I have in my process a manometer used to set up the operating presure of some vessel.

2) I have in my process a termocouple to measure the temperatue which a reaction takes place

None of them is used to measure product characteristics. They are measured used other instruments.

Am I statutory obligated to do R & R, bias, etc on these instruments ?

 

[Marc]

If the temperature is critical to the process, it will be on your process FMEA. If it is not it won't be. If your process FMEA includes a process monitoring gage, it will probably be on the control plan because it's critical to the process. If it's on the control plan it is treated as any other gage. I would expect some type of MSA study - it may not be an R&R - to validate it.

Remember that an R&R is not the only measurement study one can do and often on process gages they're not an appropriate measurement system study. There are not multiple operators (one of the "R's" in R&R) taking sequential, discrete measurements per se. It's typically a constant measurement type situation. Example - a process is temperature critical so you have a thermocouple in place. It is constantly measuring. You may even record results constantly on tape (well, data in these days) or you may even have a system where a computer reads the temp every x seconds or minutes. You have to look at each situation and determine the appropriate type of measurement study (or studies) which best suits the specifics of the situation.

And don't forget - it's measurement system analysis so if you have a thermocouple 'gage' and a computer is recording readings from it every x seconds or minutes, your study has to address the system as a whole - not just individual parts.

I would really like to hear from some others on this. I can remember back only 8 to 10 years ago I rarely saw a process monitoring gage (even a 'critical' one) even calibrated, much less any type of measurement study performed on it. QS 9000 had a lot to do with bringing the issue of process gages into the light, so to speak.

 

[Al Dyer]

Let me try this example:

I have a laser gage that measures and adjusts the orientation of 2 tool holders. We need this to ensure that only good product goes to the next process which will mill down the part to its specification.

Although the laser gage is not measuring the product, it is affecting the process and could cause defective product, crashes, unscheduled maintenance, etc...

Therefore, we do some type of MSA on the laser gage.

At the very least we do a gage R&R on every new gage that passes through the doors.

 

[Marc]

Please, please, please.... Remember what R&R standands for and the original intent was to do R&Rs with the operators who would actually use them. The reason is that if, for example, a cal lab is responsible for checking gages when they arrive (in many companies only the Calibration Manager or equivalent can even order a gage, not to mention responsiblilty for getting it checked and into the system). S/he has a few lab techs do an R&R. The bias introduced by the lab people (being 'more attuned to gages and what they're all about) will typically differ - sometimes substancially - from what would be obtained using the actual operators. This is why statements such as:
> At the very least we do a gage R&R on every new gage that
> passes through the doors.
scare me. Many times I have seen the calibration people do the R&R in the lab, come up with a great number, but when it reaches the floor the problems start. I have even seen complex gages checked in the lab and sent to the floor without training for those who have to use it.

Lastly, remember that you cannot do an R&R on most process gages because they 'become part of the process machine' so to speak. An R&R needs multiple people. As I understand it. :thedeal:

 

[Al Dyer]

Just as an aside Marc, I would never use "Lab" or "Quality" people to do an R&R study. Random operators from the floor in this case because operators are rotated between processes.

 

[Marc]

Unless the lab or quality people are the people who are going to be the folks who will be using the gage / instrument / device (whatever).... :ca:

 

[Al Dyer]

Agree, needs to be based on end usage of the gage.

 

[omolina]

Thanks to all of you!. Now I get the whole picture.
I apreciate your comments. It's time to work.....

O. MOLINA