Tolerance for Cpk purposes on certain diameters

Golfman25

Trusted Information Resource
#1
We get prints which require a Cpk on certain diameters. We usally balk at these requirements because the print tolerances are way too tight for our process to show cabability. So, I have been asked to recommend a tolerance which will allow us to guarantee a 1.67 Cpk value. What is the best way to go about determining this? Thanks in advance.
 
Elsmar Forum Sponsor

bobdoering

Stop X-bar/R Madness!!
Trusted Information Resource
#2
Re: Tolerance for Cpk purposes

What is the process you use to make the diameters? It makes a difference as to whether Cpk is even applicable.

People rubber stamp Cpk on a print when they have no idea what the distribution is of the process that generates it. That is a problem.

For customers that rubber stamp and have no clue, you can blow them off my back calculating USL and LSL out of the Cpk calculation, using an additional standard deviation to allow for long term variation:

Cpk = (USL-Xbar)/3 sigma

USL (proposed)= Xbar+(Cpk*4sigma)

Cpk=(Xbar-LSL)/3 sigma

LSL (proposed)=Xbar-(Cpk*4sigma)

Of course, it is really only applicable to centered bilateral distributions. Gets real, real ugly beyond that.

Sometimes you have to pick your battles.
 

Jim Wynne

Staff member
Admin
#3
Re: Tolerance for Cpk purposes

We get prints which require a Cpk on certain diameters. We usally balk at these requirements because the print tolerances are way too tight for our process to show cabability. So, I have been asked to recommend a tolerance which will allow us to guarantee a 1.67 Cpk value. What is the best way to go about determining this? Thanks in advance.
As Bob suggests, this is a slippery slope and you won't be able to get any traction by trying to navigate the slope backwards (i.e., setting tolerance limits and then trying to see if you can meet them) You need to know what your process is economically capable of doing before you can suggest what the proper tolerances might be. If you don't know the process, you shouldn't be making any guarantees.
 

Golfman25

Trusted Information Resource
#4
Re: Tolerance for Cpk purposes

Thanks for the quick replies. Process is drawn stamping.

We do know our process. That is the problem. What customers want is generally impossible. Of course my tooling and production people what the largest tolerance possible. I need to come somewhere in the middle -- with something that is consistently achievable. Unfortunately there are a lot of variables and I can't predict the future.
 

bobdoering

Stop X-bar/R Madness!!
Trusted Information Resource
#5
Re: Tolerance for Cpk purposes

Thanks for the quick replies. Process is drawn stamping.

We do know our process. That is the problem. What customers want is generally impossible. Of course my tooling and production people what the largest tolerance possible. I need to come somewhere in the middle -- with something that is consistently achievable. Unfortunately there are a lot of variables and I can't predict the future.
If you are doing drawn stamping, do you start with (for example) ID tooling large to allow the stamping die to wear to the small side, or is your material influence variation so large that it masks tool wear?
 
W

WKHANNA

#6
Re: Tolerance for Cpk purposes

I run into a similar situation with our biggest customer all the time.
They rubber stamp Cpk on things like ‘No Burr Allowed’ which obviously is an attribute with a Yes/No function. Try plug’n that one into MiniTab!:confused:

We also get 1.66 Minimum Cpk requirements on OD’s with +/- .001 tolerances on metal turned parts that are heat treated and plated. I guess we are supposed to be fortune tellers and oracles too!:magic:

We may at times bend the ‘rules’ a bit when calculating Cpk’s, but at the same time we have enough data on our processes that we can feel confident in our ability to deliver a product that meets the intended requirement of the customer.

Without first having a baseline or benchmark on your own processes I would be cautious.
Have you run any recent capability studies on your production?
If so, how close are you to the customers requirements?
 
G

Geoff Withnell

#7
Re: Tolerance for Cpk purposes

I think the best way to handle this sort of situation is to look at the purpose of the cpk number. It is a predicter of the expected number of defects in the output of the process, given the process spread and centering, and the given tolerance. Since most processes are NOT normal (I don't care what they told you in basic statistics class) especially around the tails of the spread, defect percentage may be a better descriptor of the expected number of defects. Using the normal curve, convert the required cpk to an expected percent defective. Look at your process history. If the historical percent defective is less than the cpk derived number, you are meeting the customer's implied need.

Geoff Withnell
 

bobdoering

Stop X-bar/R Madness!!
Trusted Information Resource
#8
Re: Tolerance for Cpk purposes

They rubber stamp Cpk on things like ‘No Burr Allowed’ which obviously is an attribute with a Yes/No function. Try plug’n that one into MiniTab!:confused:
That is a perfect example of a rubber stamp!!!!

We also get 1.66 Minimum Cpk requirements on OD’s with +/- .001 tolerances on metal turned parts that are heat treated and plated. I guess we are supposed to be fortune tellers and oracles too!:magic:
You have a better chance of calculating Cpk for the height of grass in a freshly cut lawn. The metal turned parts are easy...although Cpk does not apply if the process control is correct. Heat treat and plating, however, is a whole different story.

We may at times bend the ‘rules’ a bit when calculating Cpk’s, but at the same time we have enough data on our processes that we can feel confident in our ability to deliver a product that meets the intended requirement of the customer.
Some folks in automotive should read the WHOLE section on capability indices in the fourth edition PPAP manual (2.2.11). They would be surprised what they find! The usually stop after the first paragraph.
 
F

falconer65

#10
Re: Tolerance for Cpk purposes

I agree, the whole automotive system is fragmented.

It seems a lot of tolerances are set to the old rule that they were control limits. Parts were expected to be found out of tolerance on occasion.

Ok, in trying to make the OP some headway, does this make sense?

CP=(USL-LSL)/(6*sigma)

Since USL-LSL is my total tolerance, I can simplify the formula.

CP=(Tol)/(6*sigma)

Solving for Tol

Tol=CP*6*sigma

Desire a CP of 1.667.

Sigma is 0.1.

1.667*6*0.1=1.00

So the total tolerance would be 1.00. Since 1.667*6=10.0 take whatever the sigma is and multiply by 10.

Find the average of the process and apply the tolerance.

I don't know if this would work or not. There are so many assumptions being made. But it might be a starting place.
 
Thread starter Similar threads Forum Replies Date
D Tolerance definition based on expected Cp/cpk Reliability Analysis - Predictions, Testing and Standards 14
O Cpk for Unilateral Tolerance Capability, Accuracy and Stability - Processes, Machines, etc. 9
Hami812 Cpk Setting Tolerance - Cart before the Horse? (Wifi Routers) Capability, Accuracy and Stability - Processes, Machines, etc. 3
P Profile Tolerance and CPK Manufacturing and Related Processes 3
G Should I calculate both Cp and Cpk for a Unilateral Tolerance? Statistical Analysis Tools, Techniques and SPC 8
A How to calculate Cpk & Ppk if the tolerance changes during the period? Capability, Accuracy and Stability - Processes, Machines, etc. 5
F How to Calculate Cpk with only a Minimum Tolerance Capability, Accuracy and Stability - Processes, Machines, etc. 11
G Cpk vs. "Bad" Parts - Unreasonable Diameter Tolerance and a 1.67 Cpk Statistical Analysis Tools, Techniques and SPC 22
E Asymmetric Tolerance - How does an Asymmetric Tolerance affect the Cpk/Cp calculation Using Minitab Software 3
G Cpk Calculation - Tolerance to maintain 1.33 & 1.67 Cpk respectively Capability, Accuracy and Stability - Processes, Machines, etc. 17
S Percent of Tolerance - Cpk - Low lot sizes (40-60 pcs.) Statistical Analysis Tools, Techniques and SPC 4
L Calculating Cpk on a single sided (unilateral) tolerance Capability, Accuracy and Stability - Processes, Machines, etc. 28
S Determining Cpk on a single sided tolerance dimension Capability, Accuracy and Stability - Processes, Machines, etc. 3
D Tolerance symbol Inspection, Prints (Drawings), Testing, Sampling and Related Topics 8
D Calibration tolerance question using Pipettes Medical Device and FDA Regulations and Standards News 1
M Determining a tolerance value for Measuring devices in-house inspection General Measurement Device and Calibration Topics 12
G 0.00005" tolerance on micrometer with 0.0005" resolution? General Measurement Device and Calibration Topics 4
N Drawing tolerance vs. Measurement device General Measurement Device and Calibration Topics 7
I Control Plan (Product/Process specification/ Tolerance) acceptance FMEA and Control Plans 27
J Mechanical inspection techniques of close tolerance parts Inspection, Prints (Drawings), Testing, Sampling and Related Topics 4
Proud Liberal Small radius with tight tolerance on a torus General Measurement Device and Calibration Topics 4
L GRR for a tolerance that has changed Gage R&R (GR&R) and MSA (Measurement Systems Analysis) 6
T How to justify this High %Tolerance Gage R&R (GR&R) and MSA (Measurement Systems Analysis) 3
N % Tolerance - Type 1 study on the gages, then a gage R&R (ANOVA) Gage R&R (GR&R) and MSA (Measurement Systems Analysis) 4
S Tolerance limits for micro balance General Measurement Device and Calibration Topics 1
J Determining SPC tolerance Statistical Analysis Tools, Techniques and SPC 21
Q % Study variation low, % tolerance high - GR&R Interpretation help Gage R&R (GR&R) and MSA (Measurement Systems Analysis) 3
R Uncertainty in measurement larger than tolerance Measurement Uncertainty (MU) 2
WEAVER Electronic Weighing Scale Calibration Tolerance Manufacturing and Related Processes 1
I Determining Calibration Tolerance of a Measurement Device General Measurement Device and Calibration Topics 2
Ronen E Tolerance intervals (?) question - Flow Rate Probability Range Statistical Analysis Tools, Techniques and SPC 6
D Tolerance when verifying an analytical balance General Measurement Device and Calibration Topics 3
E Discussion between co-worker on tolerance and uncertainty and how to apply it. Thoughts? 17025 ISO 17025 related Discussions 1
C Thread Tolerance - Can I use a 3A Ring on a part that the drawing calls out for a 2A? Manufacturing and Related Processes 5
T GRR based on part tolerance or process variation. Gage R&R (GR&R) and MSA (Measurement Systems Analysis) 7
A Excel OOT (Out of Tolerance) Formula Excel .xls Spreadsheet Templates and Tools 3
N Tolerance Interval plots in Minitab Statistical Analysis Tools, Techniques and SPC 0
S Question about a basic additive variance/tolerance example Quality Tools, Improvement and Analysis 2
G Assigning a calibration tolerance - An x-y coordinate machine - Uncertainty as my verification tolerance General Measurement Device and Calibration Topics 4
R Feeler Gauge Calibration - What tolerance is used for calibrating feeler gauges? General Measurement Device and Calibration Topics 4
S SPC (Statistical Process Control) for Unilateral Tolerance - Questions Statistical Analysis Tools, Techniques and SPC 6
B Gage R&R Acceptable (10-30%), deduct Total Variation from Tolerance Gage R&R (GR&R) and MSA (Measurement Systems Analysis) 12
Y How to fix weight tolerance for plastic injection molded part? Does it vary with material groups? Manufacturing and Related Processes 1
F Position tolerance check gage manufacturing Inspection, Prints (Drawings), Testing, Sampling and Related Topics 8
M GD&T tolerance or band? Using a symbol like parallelism // Inspection, Prints (Drawings), Testing, Sampling and Related Topics 8
J What is the difference between Process Variation and Tolerance? Gage R&R (GR&R) and MSA (Measurement Systems Analysis) 4
J Gage R&R without tolerance Gage R&R (GR&R) and MSA (Measurement Systems Analysis) 8
A Calibration of Pressure Gauge, How to Determine Calibration Tolerance General Measurement Device and Calibration Topics 1
V Using K-Factor(Tolerance Interval) Analysis for Design Verification Statistical Analysis Tools, Techniques and SPC 3
C Acceptable NDC for %GR&R part inspection to Tolerance (%Tolerance) Gage R&R (GR&R) and MSA (Measurement Systems Analysis) 3

Similar threads

Top Bottom