Revised Date: 12/08/04 MAGNA POWERTRAIN NEW PROCESS GEAR INC. SPECIFICATION NO.

109

1 of 11 Pages

STATISTICAL REQUIREMENTS FOR GAGES - R & R

After the gage is built, gage must pass R&R in NPG's Tool & Gage Department. See pages 5 & 6 for procedure of R&R. Ref. Measurement System Analysis (MSA Feb. 1995) When the Gage Supplier performs a gage R&R on their premises, the Gage Supplier must send the R&R study in with the gage. Note: NPG shall absorb the cost of supplying the parts; the Gage Supplier shall absorb the investment in conducting the R&R. If parts are not available (i.e. new product launch), a Gage accuracy study performed on a Part-simulated master can reveal weakness in design or build if average exceeds 10% of part print tolerance. Simulating gage use at the vendor's location can only help the Gage Supplier reach the goal of Zero Defects.

CRITERIA FOR ACCEPTANCE - R&R

New Process Gear as purchaser and user of gages seeks to buy the best possible equipment to measure our components. The gage should be accurate, stable, and linear. The R & R study shall use a minimum of 2 operators with 2 trials per operator on 10 production parts.

Revised Date: 12/08/04 MAGNA POWERTRAIN NEW PROCESS GEAR INC. SPECIFICATION NO. 109

2 of 11 Pages

CRITERIA FOR ACCEPTANCE - R&R

The criteria for acceptance of the gage are: R&R less than or equal to 10% error - Acceptable. R&R of 10% to 30% error - May be acceptable with appropriate waiver letter. The signatures of NPG's Manager of Manufacturing Engineering and Manager of Quality Systems and Services are required on the waiver letter. R&R over 30% - Not Acceptable. * The final decision to accept/reject a gage shall be based on the NPG performed Gage Studies. The R&R requires the tolerance of the part dimension to be measured. NPG performed Gage Studies. The R&R requires the tolerance of the part dimension to be measured. NPG will provide the part dimension tolerance for this purpose. All questions should be referred to gage requestor. NOTE TO GAGE SUPPLIER: Gage received with excessive R&R will be rejected and returned to supplier at Gage Supplier's expense.

* See page10 "NOTE"

Revised Date: 12/08/04 MAGNA POWERTRAIN NEW PROCESS GEAR INC. SPECIFICATION NO. 109

3 of 11 Pages

GLOSSARY OF TERMS FOR GAGE R&R

Gage R&R - Gage Repeatability and Reproducibility Gage Repeatability - Gage Repeatability is the variation in measurements obtained when one operator uses the same gage for measuring the identical characteristics of the same parts. Gage Reproducibility - Gage Reproducibility is the variation in the average of measurements made by different operators using the same gage when measuring identical characteristics of the same parts. Gage Accuracy - The difference between the observed average of measurements and the true average of the same parts using precision instruments. Gage Linearity - The difference in the accuracy values through the expected operating range of the gage. Gage Precision - See Gage Repeatability Gage Stability - The difference in the average of at least two sets of measurements obtained with a gage on the same parts as a result of time. Gage System Error - The combination of gage accuracy, repeatability, reproducibility, stability and linearity.

SYMBOLS AND ABBREVIATIONS USED IN THIS ATTACHMENT

3 s GRR R R X = X

Sum or Total Sample Standard Deviations Gage Repeatability and Reproducibility (Gage Error) Range Average Range Average of Measurements Grand Average of Several Series of Measurements

4 of 11 Pages

Operator/Trial # 1 1. 2. 3. 4. 5. 6. 7. 8. 9.
10. 11. 12. 13. 14. AVG.

Part 2 3 4 5 6 7 8 9 10

Average

1 2 3

AVG. RNG. B 1 2 3 AVG.

RNG. C 1 2 3

Xa = Ra =

Xb =
Rb =

Xc= Rc = Rp = +Rc = ] / [ # OF OPERATORS = ] = R=

15. 16.

RNG. PART AVG (Xp)

17. [Ra = +Rb = 18. [ MaxX= 19. [ R =

Min X = ] = X Diff x D4 = ] = UCLR

20. [ R =

x D3 =

] = LCLR

*D4 = 3.27 for 2 trials and 2.58 for 3 trials. UCLR represents the limit of individual R's. Circle those that are beyond this limit. Identify the cause and correct. Repeat these readings using the same appraiser and unit as originally used or discard values and re-average and recompute R and the limiting value from the remaining observations. Notes:

5 of 11 Pages

Part No. & Name: Characteristics: Specification:

Gage Name: Gage No: Gage Type:

Date: Performed by:

From data sheet:

X Diff

Rp = % Process Variation
% EV = 100 [EV/TV] = 100 [ / ] % =

Measurement Unit Analysis

Repeatability - Equipment Variation (EV)

EV

= R x K1 = x =

Trials 2 3

K1 4.56 3.05 % AV = 100 [AV / TV] = 100 [ / ] [ / ] = % =

Reproducibility - Appraiser Variation AV = p [ (x Diff x K2)5- (EV5/nr)] = p [ ( X )5- ( 5 / x )]

Operators K2 Repeatability & Reproducibility (R &R) R&R = p[( X = p[( X =

2 3.65

3 2.7

n = number of parts r = number of trials

)5- ( 5 / x )] )5- ( 5 / x )] Parts K3

% R & R = 100 [ R & R / TV] = 100 [ / = %

Part Variation (PV) PV = = = Total Variation (TV) TV = p (R&R5 + PV5) = p ( 5 + 5) =

All calculations are based upon predicting 5.15 sigma (99.0% of the area under the normal distribution curve). K1 is 5.15 d2, where d2 is dependent on the number of trials (m) and the number of parts times the number of operators (g) which is assumed to be greater than 15. AV - If a negative value is calculated under the square root sign, the appraiser variation (AV defaults to zero (0). K2 is 5.15/d2, where d2 is dependent on the number of operators (m) and (g) is 1, since there is only one range calculation. K3 is 5.15/d2, where d2 is dependent on the number of parts (m) and (g) is 1, since there is only one range calculation. d2 is obtained from Table D3, "Quality Control and Industrial Statistics". A.J. Duncan.

Rp x K3 x

2 3 4 5 6 7 8 9 10

3.65 2.70 2.30 2.08 1.93 1.82 1.74 1.67 1.62

% PV

= 100 [ PV / TV ] = 100 [ = / % ]

Revised Date: 12/08/04 MAGNA POWERTRAIN NEW PROCESS GEAR INC. SPECIFICATION NO. 109 1. HOW TO PREPARE FOR A MEASUREMENT SYSTEM STUDY

6 of 11 Pages

The first step in preparing for a Measurement System Study is establishing the purpose of the study and determining the kind of information needed to satisfy the requirements. The following questions should be answered: What approach should be used? See 2, 3 below. How many operators are to be involved? Minimum of (2) - (3) preferred. How many sample parts are to be tested? Ten What number of repeat readings will be needed? Short form acceptable?

Once that information is outlined, the next step is to choose the operators and select the sample parts to be used for the study, etc.

2.

VARIABLE GAGE STUDY FOR REPEATABILITY AND REPRODUCIBILITY The variable gage study can determine errors of gage repeatability and reproducibility (R&R) separately. Study results can also provide information concerning the causes of gage error. For example, if lack of reproducibility is large compared to repeatability, then possible causes could be: a. b. The operator is not properly trained in how to use and read the gage instrument. Calibrations on the gage dial are not clear.

On the other hand, if lack of repeatability is large compared to reproducibility, the reasons may be: a. b. c. The gage instrument needs maintenance. The gage should be redesigned to be more rigid. The clamping or location for gaging needs to be improved.

3.

CONDUCTING THE STUDY Even though the number of operators, the number of trials and the number of parts may be varied, the study consists of the following steps: a. b. c. d. Refer to operators A, B and C and number the parts 1 through 10 so that the numbers are not visible to the operators. Calibrate the gage. Let Operator A measure 10 parts in a random order and enter the results in Column 1 on Page 4. Repeat step 3 with Operators B and C.

Revised Date: 12/08/04 MAGNA POWERTRAIN NEW PROCESS GEAR INC. SPECIFICATION NO. 109 3. CONDUCTING THE STUDY (continued) e. f.

7 of 11 Pages

Repeat the cycle, with the 10 parts measured in another random order, for the number of trials required. Steps c through e may be modified for large size parts, unavailability of parts or when operators are on different shifts. Using the attached form, enter the observations on one side (Page 4) and calculate Gage R&R using the formulas shown on the reverse side (Page 5).

g.

Note: Measurements for the study should be taken in random order in prevent bias.

4.

GAGE ACCURACY *

Lack of gage accuracy generally is not as predominant a gage error as repeatability and reproducibility. Nevertheless, it should be identified. To determine gage accuracy, it is necessary to obtain the true measurement of the sample parts. This can usually be done with tool room or layout inspection equipment. A true average is derived from these readings to be compared later with the operator's observed averages (identified as X, X, X, ) from the gage R&R study.

If it is not possible to measure all of the sample parts in this manner, the following alternative method can be used. Measure one of the sample parts precisely on tool room or layout inspection equipment. Have one operator measure the same part a minimum of 10 times, using the gage being evaluated. The difference between the true measurement and the observed averages represents gage accuracy. To convert accuracy to a percentage of tolerance, multiply by 100 and divide by the tolerance. (Must not exceed 10% of part tolerance). If gage accuracy is relatively large, look for these possible causes: a. b. c. d. e. f. Error in the master Worn gage Gage made to the wrong dimension Gage is measuring the wrong characteristic Gage not calibrated properly Gage being used improperly by operator

* Generally done with masters if part simulated.

Revised Date: 12/08/04 MAGNA POWERTRAIN NEW PROCESS GEAR INC. SPECIFICATION NO. 109 5. GAGE STABILITY

8 of 11 Pages

Gage stability should be computed although it typically is not as large as repeatability and reproducibility. How gage stability is determined depends upon how often the gage is used between normal calibrations. The need for calibration results from factors such as these: Time - long idle periods Number of measurements taken Air pressure change Warm - up

Given that any of these factors are known, the number of times one must calibrate the equipment can be established to help minimize gage error due to lack of stability. If gage is used intermittently, then its stability may be determined at the same time as the gage R&R study is made. The gage must be calibrated before and after each trial to determine the amount of calibration change. This difference is the gage stability for THAT TRIAL ONLY. To calculate overall gage stability, the calibration change for each trial must be added up and divided by the number of trials: sum of calibration changes gage stability = number of trials If a gage is normally used for relatively long periods of time without calibration (such as start of a shift or after lunch) the stability can be determined without recalibrating for each trial. One would simply conduct another gage R&R study just prior to the time recalibration is due. Gage stability would be the positive difference between the grand averages calculated using all of the measurements in the first and second studies. Gage stability = range of (grand average study one - grand average study two) To convert gage stability to a percentage of tolerance, multiply by 100 and divide by the tolerance. If gage stability error is relatively large, look for these possible causes: a. b. c. Gage not being calibrated as frequently as needed. Air pressure regulator or filter may be needed for air gaging. Warm-up may be required for electronic gages.

Revised Date: 12/08/04 MAGNA POWERTRAIN NEW PROCESS GEAR INC. SPECIFICATION NO. 109 6. GAGE LINEARITY

9 of 11 Pages

Gage Linearity can be determined by conducting gage accuracy studies through the expected operating range. A minimum of two accuracy studies should be conducted, one at each end of the operating range. Gage Linearity = largest accuracy value - smallest accuracy value. If gage linearity error is relatively large, look for these possible causes: a. b. c. d. Gage not calibrated properly at both lower and upper end of the operating range. Error in the minimum or maximum master. Worn gage May require reviewing the internal gage design characteristics.

7.

DEFINITIONS Builder/Distributor/Importer/MTB authorized agent.

GAGE SUPPLIER:

MTB:

Machine Tool Builder. When Design and Build of gages is part of the Machine Tool order. i.e. responsibility for "Turn Key"/ "Black Box" operation.

Refer to NPG'S M.E. #103 - Design & Build of Tooling, Gages, & Fixtures.

8.

GLOSSARY -

See Page 3

9.

REFERENCES

Statistical Quality Control Handbook, Western Electric, Select Code 700-444, PO Box 19901, Indianapolis, Ind. 46219, (800) 432-6000 Measurement System Analysis, Chevrolet Motor Division, GMC. "Measuring Equipment Repeatability - The Rubber Ruler," Robert Travor, 1962 ASQC Convention Transactions. "Industrial Quality Control," H.C. Charbonneau and G.L. Webster, Prentice-Hall., NY 1978.

Revised Date: 12/08/04 MAGNA POWERTRAIN NEW PROCESS GEAR INC. SPECIFICATION NO. 109 9. REFERENCES (continued)

10 of 11 Pages

"Measurement System Analysis" (Feb 1995) GENERAL MOTORS STATISTICAL PROCESS CONTROL MANUAL. "NOTE" NPG reserves the right to utilize the AT & T Measurement Error Study to analyze gages who's R&R exceeds 30%. (See pp. 87-89 in AT&T SQC Handbook, 1956). Criteria for using the AT & T Measuring Error Study 1. A particular product has a Cpk less than 1.00; practically speaking there is a high percentage of defects. Process improvement is obviously needed, Gage is tested on R & R and has greater than 30% variation. A temporary gage is not available to sort or measure parts. NPG is experiencing heavy cost penalties due to the lack of an adequate measuring device: sorting, tear-downs, repairs, scrap and customer complaints.

2. 3. 4.

If these elements are present, the investment it a 40 part Measurement Study probably has merit. The objective of the study is to answer the following: Where do resources need to be allocated to improve our process? a. b. Time and $ spent to improve the gage? or Time and $ spent working on the product?

Some gages with R&R's above 30% have been accepted for use on problem processes because the ratio of gage variation to part variation is small enough to discern variation in the product. Acceptance of a gage using this method is at the discretion of NPG Gage Engineer. When the product is improved to a level where the gage needs improving, a new gage will need to be requisitioned. Summary of Note: The AT&T Measurement Error Study will hold precedence over the GR & R method in cases where vendors supply non-conforming material. Use of this method will be solely at the discretion of NPG.

Revised Date: 12/08/04 MAGNA POWERTRAIN NEW PROCESS GEAR INC. SPECIFICATION NO. 109 CONDITIONS OF ACCEPTANCE PART NO: PART NAME:

11 of 11 Pages

PART DIMENSION: PART TOLERANCE:

Gage supplier agrees to a 10 piece Repeatability & Reproducibility Test at our plant per M.E. Spec. #109.

GAGE SUPPLIER:

MFG. ENG'R: QUALITY ENG'R:

DATE: DATE: