Achieving cpk of 1.33
- Thread starter Brik1979
- Start date
What do you mean by UCL and LCL? These are typically the upper control limit and lower control limit for SPC. Then there are upper and lower specification limits (USL and LSL) Can you clarify what you mean?
Also I’m not sure what you are asking when when you say “how do I achieve all dimensions above 1.33”. Are you asking for advice on how to tighten up the results on a turning operation? Golfman has some good starting device here. That might not be possible if you don’t know the dominant cause of variation in your processes. But to help you more than that we’d need to know what kind of lathes you have, how they are controlled, what kind of metal you are cutting, how are dimensions measured in the lathe and out of the lathe, have you done an MSA? And most importantly we’d need to see a multi-vari chart of your results…(google it)
I had two different sized parts, same metal (super alloy), multiple cuts to achieve the final dimension and each part had a MSA performed. The cause of variation for each was quite different. One met the specs easily but had a very large SD due to lot to lot variation which dominated the total variation. The other part didn't meet the specifications and the Sd was dominated by the within piece variation. The actions to get to a 1.33 Cpk were quite different for these parts…
Also I’m not sure what you are asking when when you say “how do I achieve all dimensions above 1.33”. Are you asking for advice on how to tighten up the results on a turning operation? Golfman has some good starting device here. That might not be possible if you don’t know the dominant cause of variation in your processes. But to help you more than that we’d need to know what kind of lathes you have, how they are controlled, what kind of metal you are cutting, how are dimensions measured in the lathe and out of the lathe, have you done an MSA? And most importantly we’d need to see a multi-vari chart of your results…(google it)
I had two different sized parts, same metal (super alloy), multiple cuts to achieve the final dimension and each part had a MSA performed. The cause of variation for each was quite different. One met the specs easily but had a very large SD due to lot to lot variation which dominated the total variation. The other part didn't meet the specifications and the Sd was dominated by the within piece variation. The actions to get to a 1.33 Cpk were quite different for these parts…
Thank you for your valuable advice. My dimensions are falling under range but are not centered. we are using CNC turning with SAE 1010 & using collet chucks for clamping & all finishing operations including all sc's in one operation. we are using mechanical spin type comparators, bore guage for OD & ID inspections.
what can be possible sources of variation to look at Machine & Process?
what can be possible sources of variation to look at Machine & Process?
What is your tolerance? I am doubtful a standard bore gage will get you there.
Possible sources of variation can be grouped into large families that hold al specific factors. These groupings are called components of variation or families of variation. To begin you do not need to know every single specific factor - there are undress if not thousands of them. The key is to begin thinking of how the characteristic varies. The components are:
Measurement error
Within piece
Piece to piece
Lot to lot
Time to time. (Time to time can be long or short: Week to week, cutter to cutter, cut to cut, adjustment to adjustment)
Operator to operator
Machine to machine
each of the families are mutually exclusive and exhaustive.
Plot your data on a multi-vari chart (again do your research and look it up.)
I will try to post a presentation I did on this at a later date (It is the Thanksgiving holiday)
Measurement error
Within piece
Piece to piece
Lot to lot
Time to time. (Time to time can be long or short: Week to week, cutter to cutter, cut to cut, adjustment to adjustment)
Operator to operator
Machine to machine
each of the families are mutually exclusive and exhaustive.
Plot your data on a multi-vari chart (again do your research and look it up.)
I will try to post a presentation I did on this at a later date (It is the Thanksgiving holiday)
50 microns
Is that total tolerance? 50 microns is about .002 total. If +/- .002 (50 microns) you have a chance. If +/- .001 (25 microns) you're going to have to look at everything -- setups, cucks, tooling, measuring equipment, etc. Can you check the part on on CMM?
Now you say your dim is not centered. Can you center it?
50 microns? Best to start with an MSA…
Similar threads
