Southwest Test Conference - SWTest.org

1 Introduction toGage R&R StudiesThe Key to UnderstandingMeasurement SystemsByHank Scutoski & Chander Sekar, CORPORATIONG ilbert, ArizonaSouthwest Test Conference Speakersq Hank Scutoski VP Quality 30 Years Experience TI, Motorola, Cerprobe Motorola University s Six Sigma Research Institute Certified Quality Engineer Certified Quality Auditorq Chander Sekar, Corporate Statistician 23 Years Experience Professor of Statistics at the University of Madras, India 3 Degrees in Statistics Certified Quality Engineer We must determine the uncertainty ofour measurement systems before wecan compare, control or optimize ourmanufacturing s Philosophy Studies Completed85 Studies to Date Including: Tip Diameter Tip Length Contact Force Probe Tip Alignment Planarity Contact Resistance Leakage Benefitsn Validates consistent results between:q Cerprobe s many manufacturing facilitiesq Cerprobe & customer sitesn Provides for Dock to Stock certificationn Provides feedback to Cerprobe s Able to provide customer with quality products with statistical consulting support.

2 Statistical ApplicationsGage R&RGage R&RScientificGuard BandingScientificGuard BandingComparison ofMetrology ToolsComparison ofMetrology ToolsSPCSPCD esign ofExperimentsDesign ofExperiments PrecisionR = Diameter = 6 Std DevTrue ValueX1X2X3 The variation obtainedfrom one gage and oneoperator whenmeasuring the samepart several times. Machine Variation Only Applies to ATEGage Repeatability Gage ReproducibilityTrue ValueX1X2X3Y1Y3Y2 Precision R&R The difference in theaverage of themeasurements made bydifferent operators usingthe same gage whenmeasuring the samepart. Operator-to-Operator Variation R&R Summary Plot Operators of Operators:3No. of Parts:32No. of Trials:3 Deviation from AverageMicroscope Operators from AverageNo. of Operators:3No. of Parts:32No. of Trials:3 Study #3R&R Summary Plot Video System Total Variationq Process Variationq Gage Variation Repeatability Reproducibility Note:q One Gageq 1 Operatorq Repeated Measurements o o o oooo ooo o ooo o-3 +3 ReproducibilityOperator # 2 Distribution Operator # 1 DistributionNote:q One Gageq 2 Operatorsq Repeated Measurementsq One Part Measurement System Capabilityn Precision-to-Tolerance Ratio (P/T) P/T < Acceptable System < P/T < Marginally Acceptable System P/T > Unacceptable SystemRef.

3 : SEMATECH: Introduction to Measurement Capability Analysis #91090709A-ENG Interpretation of P/T RatiosP/T = * Measurement system consumes 10% of tolerance * Effect of T on P/T P = = T = = P/T = P/T = Repeatability Studyn Single operator performs multiple trialsn No change in the setup between trialsn Provides a quick estimate of measurement capabilityn PR/T Sample data on Align (x) from a repeatability study Trial #1 Trial #2 Trial #3 Trial #4 Trial #5 S :Standard Deviation R: RangeRepeatability Study - Data Repeatability StudyP/T Calculations PR/T = 6 * ( R / d2) / (USL - LSL)_ Gage R&R Studyn Involves multiple operators and trialsn Total tear down of the setup between trialsn Provides separate estimates of repeatability and reproducibilityn PR&R /T Sample data on Align (x) from a full gage R & R study O1-T1 O1-T2 O1-T3 O2-T1 O2-T2 , O2 - Operators 1 and 2T1, T2, T3 - Trials 1, 2 and 3 Gage R&R Study - Data Gage R&R Studyn Repeatabilityn Reproducibilityn Operator-Part Interactionn Part-to-Part Variation SignificantOperator-to-PartInteractionSi gnificantOperator-to-PartInteractionBad GageBad GagePoor OperatorTrainingPoor OperatorTrainingImproperMeasurementStudy ProcedureImproperMeasurementStudy ProcedureOperator-to-Part Interaction Data Analysis Methodsn Average Range Methodn Analysis of Variance (ANOVA)

4 Method P/T Calculations PR&R/T = 6 * ( R&R Sigma) / (USL - LSL)Gage R&R Study Road Map to with SuppliersFeedbacktoOperatorsRobustDataCo llectionAdvancedStatisticalTechniquesCon sistent Product QualityConsistent Product Quality