Process Capability in a High Precision Environment

[Marlin01]

Hi Everyone,
I just started introducing the automotive process control approach in a non-automotive, high precision manufacturing business. Immediately it's clear that many of the grinding tolerances (0.003mm) are very difficult to achieve and even harder to measure due to resolution issues. Just capturing the data to calculate capability to Cpk1.69 is proving impossible.
Has anyone out there been in this position, what are the options? Is there another control option other than using capability?

 

[bobdoering]

Yes, grinding is the gold standard of the precision machining process control! First, you need to recognize that your capability is a function of what two points you adjust between. Since grinding creates a sawtooth curve - wear, adjust, wear, adjust - it is non-normal and the standard capability calculations DO NOT APPLY! They assume normality and random variation. Wheel wear is not random, it is a function over time. I recommend reading through the blogs on precision machining process control here, and ask any questions you might have afterwards. The wheel has already been invented!

As far as gaging, yes a 3 micron total tolerance is very tight. You need a gaging system that goes down to 0.1 micron (air gage?) AND you need to recognize if that tolerance is real you are in the realm of pristinely clean parts being measured at a temperature controlled environment - or you are kidding yourself. Your control limits to meet a 1.33 capability is +/- 1.1 micron.

 

[optomist1]

My two cents, amen to Bob's statement re: precision grinding and precision machining not following the normal distribution.....it took me awhile to get me head around this concept.

 

[Marlin01]

Hi Guys;
Thanks for your responses. I was sure that there were people out there that had faced this issue before. I can confirm gladly that you have immediately grasped the exact point I was hoping to address. Bob, I will follow up on the links your sent and come back later with questions if I have any.

I can confirm that the tolerances are real, at least our engineers believe they are required! I can also confirm that we are using air gauges and controlling the temp. The team is technically competent, they have simply never before faced a requirement for Special Characteristics and the related capability requirements.

 

[Marlin01]

HI BobDoering;
The link you posted doesn't lead anywhere, could you try again please?

Thanks.

 

[Bev D]

a good research effort or study effort (like a good experiment) seeks to challenge our thinking rather than to confirm it...in this light let me further challenge you on the need to use capability indexes especially the AIAG approach. are you doing this because a Customer is requiring it or are you doing it because you think it might be a "good thing"?

Too often we take on 'popular' approaches without critical thought as to their validity or usefulness. As my Mother constantly reminded me: "just because all of the other kids are doing it doesn't mean you should do it" followed by "if all the other kids jumped off a bridge would you?"

 

[Miner]

HI BobDoering;
The link you posted doesn't lead anywhere, could you try again please?

Thanks.

I just verified that the link does work.

 

[Miner]

I can also confirm that we are using air gauges and controlling the temp. The team is technically competent, they have simply never before faced a requirement for Special Characteristics and the related capability requirements.

Don't overlook performing a measurement systems analysis on your air gauging system.

Here is a link to my MSA blogs. If you don't have a customer dictating the methodology, I recommend Wheeler's approach.