Metrology Equipment Selection for Measuring the Material Thickness of Company XYZ s Next Generation JB3 Titanium Cathode Material

Transcription

1 1 Author: Title: Kotyk, Brian, K Metrology Equipment Selection for Measuring the Material Thickness of Company XYZ s Next Generation JB3 Titanium Cathode Material The accompanying research report is submitted to the University of Wisconsin-Stout, Graduate School in partial completion of the requirements for the Graduate Degree/ Major: Research Adviser: Diane Olson, Ph.D. Submission Term/Year: Fall, 2011 Number of Pages: 72 MS Technology Management Style Manual Used: American Psychological Association, 6 th edition I understand that this research report must be officially approved by the Graduate School and that an electronic copy of the approved version will be made available through the University Library website I attest that the research report is my original work (that any copyrightable materials have been used with the permission of the original authors), and as such, it is automatically protected by the laws, rules, and regulations of the U.S. Copyright Office. STUDENT S NAME: Brian Kotyk STUDENT S SIGNATURE: DATE: Dec 19, 2011 ADVISER S NAME Dr. Diane Olson ADVISER S SIGNATURE: DATE: Dec 20, This section to be completed by the Graduate School This final research report has been approved by the Graduate School. (Director, Office of Graduate Studies) (Date)

2 2 Kotyk, Brian K. Metrology Equipment Selection for Measuring the Material Thickness of Company XYZ s Next Generation JB3 Titanium Cathode Material Abstract The purpose of this study is to identify a measurement device and process that is capable of accurately measuring the overall thickness of thin metal foils with high surface finishes. Company XYZ has specified the JB3 titanium cathode material which is grade two titanium and / in thickness with a 23 Ra (roughness average) minimum. This foil had never been manufactured prior to Company XYZ contracting to have Ullegheny who is a world leader in titanium foil manufacturing. Ullegheny currently does not have the capability to measure the JB3 titanium cathode overall thickness per Company XYZ s specifications because their current measurement system does not have the capability. In this study two devices will be evaluated as well as Ullegheny s current measurement device. The evaluations will consist of a Gage Repeatability and Reproducibility, capability study, and measurement characteristic analysis to determine the optimal device for Ullegheny to implement to measure the JB3 titanium cathode material thickness.

3 3 The Graduate School University of Wisconsin Stout Menomonie, WI Acknowledgments I would like to thank everybody at Company XYZ for being supportive of my research project and assisting me through the process. I would also like to thank Ullegheny for assisting me with the testing at their facility. My instructor Dr. Diane Olson did a wonderful job mentoring me through the research paper and I could not thank her more. Last but not least I would like to thank my wonderful girlfriend Marissa for keeping me motivated throughout the long process of research and the writing process.

4 4 Table of Contents... Page Abstract...2 List of Tables...6 List of Figures...7 Chapter I: Introduction...8 Statement of the Problem...11 Purpose of the Study...12 Assumptions of the Study...12 Definition of Terms...12 Limitations of the Study...16 Methodology...16 Chapter II: Literature Review...17 Quality Costs...17 Measurement Devices...18 Measurement Process...21 Measurement Variation...23 Concept of Gage R&R Study...29 Process Capability, Control Charts, and Statistical Tools...31 Chapter III: Methodology...33 Measurement Device Selection and Descriptions...33 Titanium Cathode Subject Selection and Description...39 Gage Reproducibility and Repeatability Test Description...40

5 5 Capability Study Description...40 Gage R&R and Capability Study Data Analysis...41 Quantitative Analysis of the Measurement System...42 Limitations...43 Chapter IV: Results...44 Statistical Analysis: Gage R&R and Capability Study...44 Quantitative Analysis of the Measurement System Results...47 Chapter V: Discussion...53 Measurement Device Selection...54 Limitations...55 Conclusions...55 References...57 Appendix A: An Introduction to APA Style. Research Paper FAQS; Provided here for your reference only; don t include in your paper...58 Appendix B: Gage R&R Raw and Analysis Data...61 Appendix C: Heidenhain work instructions for Ullegheny Ludlum for measuring the JB3 Company XYZ titanium cathode material thickness...67

6 6 List of Tables Table 1: Statistical Analysis Studies Gage R&R and Capability Study 42 Table 2: Quantitative Analysis of the Measurement System. 43 Table 3: Statistical Analysis Studies Gage R&R and Capability Study Results 45 Table 4: Quantitative Analysis Results of the Measurement Systems...47

7 7 List of Figures Figure 1: Digital hand micrometer from Company XYZ s Corlab...19 Figure 2: Direct Computer Controlled Coordinate Measuring Machine..20 Figure 3: Heidenhain CT6001 precision height measurement system at Ullegheny. 21 Figure 4: The relationship between total, process and measuring variation.. 24 Figure 5: Measurement Bias.. 25 Figure 6: Measurement Stability Figure 7: Measurement Linearity Figure 8: Measurement repeatability Figure 9: Measurement reproducibility Figure 10: Breakdown of overall variation Figure 11: This figure is showing the percent tolerance calculation Figure 12: Pp and Ppk formulas Figure 13: Vollmer VMF1000 Measurement System at Ullegheny..34 Figure 14: Vollmer VMF1000 contact sphere points Figure 15: Fowler THV Measurement system at Company XYZ s Corlab Figure 16: Fowler THV Flat Anvil Contact Surfaces Figure 17: Heidenhain CT6001 Measurement system Figure 18: Heidenhain CT6001 Flat Anvil Contact Surface...39 Figure 19: Capability study for the three measurement devices evaluated Figure 20: Capability study deviation chart for each of the measurement systems

8 8 Chapter I: Introduction Company XYZ is an industry leader in producing human implantable defibrillators. Implantable defibrillators are small battery powered pulse generator devices that deliver therapy to the heart when it senses abnormal activity. The device is programmed to monitor heart rate for rhythm abnormalities and when they occur the device delivers therapy. The device consists of a pulse generator which holds the computer board, battery, and capacitors. Connecting the pulse generator to the heart are the lead wires, which transfer the energy and signals to and from the pulse generator unit. When therapy is needed the battery charges up the capacitors and the energy jolt is delivered through the leads to the heart. The jolt resynchronizes the heart to a normal beat. There are two forms of implantable defibrillators on the market today which treat heart complications. The CRT-D (Cardiac Resynchronization Therapy Defibrillator) device treats patients with heart failure and need biventricular pacing. The ICD (Implantable Cardioverter Defibrillator) device treats patients with sudden cardiac arrest due to ventricular fibrillation and ventricular tachycardia. One of the main components in the defibrillator that facilitates the delivery of the therapy is the pair of capacitors in the pulse generator device. Company XYZ designs and manufactures the capacitors at their Minnesota campus. There are two main materials in the capacitor which are the aluminum anode and titanium cathode foils. Both materials are rolled to the desired thickness at Ullegheny, which is a metal rolling supplier for Company XYZ. Ullegheny uses a Z-Mill to roll the material to the specified thickness, width, and finish. The Z-Mill is a metal rolling machine that operates with very small diameter work rolls with high pressure to reduce the metal material thickness with pressure and tension. Once the processing of the material has been completed the metal is spooled on a coil and shipped to Company XYZ. Once Company

9 9 XYZ receives the material at the capacitor manufacturing facility, the spools of metal are utilized in a metal stamping press to produce the anode and cathode coupons which make up the capacitor. With each device generation, the capacitor requirements are to be smaller and deliver the same or better energy output as the previous generation. Therefore, the tolerances of all the components in the pulse generator tighten with each new generation. The layers of anodes, cathodes, and paper insulators in the capacitor are held to a very tight tolerance because the capacitor tolerance stack up does not allow excessive variation. If there is too much material variation and the layers stack height is too tall, the can lid will not close. If the layers are not thick enough, usually the capacitance requirement will not be met. The titanium cathode function in the capacitor is to hold the energy with capacitance based off the total surface area. The current generation capacitor is called the JB2 and the cathode has an overall thickness specification of / with a surface finish of 10 +/- 2 Ra. Company XYZ s next generation capacitor called the JB3 which is smaller in volume but still has the same capacitance as the current JB2. To achieve this requirement the engineers at Company XYZ increased the JB3 titanium cathode surface area without increasing the overall thickness. Through prototype testing the optimal JB3 titanium cathode specification was determined to have a thickness of / with a surface finish of 23 Ra minimum. The increased Ra specification from the JB2 to the JB3 significantly increased the surface area which helps achieve the capacitance requirement. The JB2 titanium cathode material with a surface finish of 10 Ra has high sheen. The JB3 is a very dull textured surface when viewed under a microscope.

10 10 In the early development of the JB3 material, there was a correlation issue of the material thickness between Company XYZ and Ullegheny who supplied the titanium cathode material. Ullegheny was measuring the JB3 material thickness with the same measurement device and following the same procedure to measuring the JB2 material but Company XYZ noticed a significant measurement shift. Ullegheny is currently using a Vollmer VMF 1000 precision height gauge to measure the material thickness. The Vollmer device is a vertical precision measurement system that has two.500 spheres as the contact points to measure the material thickness. The industry standard for measuring sheet metal material thickness at sheet metal manufactures is the Vollmer contact and X-ray measurement devices. Company XYZ s Corlab (Corporate Laboratory) has a Fowler THV horizontal precision measurement system. This is the system engineers use to measure the JB3 titanium cathode in development. It has two flat anvil.250 diameter posts that contact the material to measure the overall thickness. The major difference between the Fowler THV and the Heidenhain CT6001 is that the Vollmer has spheres that contact the surface and the Fowler has round flat surfaces. With the JB2 material Company XYZ and Ullegheny didn t have correlation issues because the material is extremely flat and has a very smooth high sheen finish. The JB3 material has a very high surface finish which is very dull with an orange peel effect. The orange peel effect gives the material a slight texture which is inherent from the high finish application. The correlation issue comes from the two different measurement methods of sphere and flat contact methods. When measuring the JB3 titanium material with the Vollmer it measures the base material thickness and the Fowler measures the overall thickness and accounts for the material texture. Thus the Vollmer measurements compared to the Fowler are consistently thinner.

11 11 Since Company XYZ is concerned with the overall material thickness, the Vollmer in its current configuration cannot accurately measure thickness to the specified requirement. Also, the sphere contact points on the Vollmer measurement device are not removable, which would have been ideal to insert flat anvils to replicate the THV measurement contact surfaces. For Ullegheny to accurately produce the JB3 titanium cathode per Company XYZ s requirements they needed to implement a measurement device capable of measuring the overall thickness specification. Two measurement systems were identified that would be able to measure the JB3 titanium cathode material thickness accurately and with a degree of repeatability. The two systems identified are the Heidenhain CT6001 and the Fowler THV measurement devices. These systems were designed to measure heights to a very high resolution which would encompass the JB3 material thickness specification callout. According to the manufacturer specifications the three measurement devices are accurate enough to measure the JB3 material thickness total tolerance of Statement of the Problem The problem is the industry standard measurement devices for measuring material thicknesses at sheet metal manufacturers do not have the accuracy to measure extremely thin, high surface finish, and tight tolerance metals. This study will evaluate different measurement devices and establish a technique to accurately measure the thin metal materials.

12 12 Purpose of the Study The titanium cathode in the next generation JB3 capacitor at Company XYZ has a thickness specification of / with a surface finish of 23 Ra minimum. Ullegheny has the current industry standard measurement device, which is not capable of accurately measuring the overall thickness of the JB3 titanium cathode material. The purpose of the study is to identify a metrology device and method to measure thin metal material with extremely tight tolerances. Assumptions of the Study 1. The metrology equipment was calibrated to its manufactures standards. 2. The data was measured by an engineer that is trained on the measurement device. 3. The data is collected in an environment that was constant to limit the variables such as temperature and humidity. 4. The devices that are to be evaluated need to be affordable for a raw material supplier, and the method of measurement needs to be efficient for manufacturing. 5. The only contributions to measurement variation are the operator and the measurement device. Definition of Terms Accuracy - The closeness of agreement between an observation value and the accepted reference value (MSA, 2010). Bias - The difference between the observed average of measurements (trials under repeatability conditions) and a reference value; historically referred to as accuracy. Bias is evaluated and expressed at a single point within the operating range of the measurement system. This also can be systematic error favoring a particular result (MSA, 2010).

13 13 Calibration A set of operations that establish, under specified conditions, the relationship between a measuring device and a traceable standard of known reference value and uncertainty. Calibration may also include steps to detect, correlate, report, or eliminate by adjusting any discrepency in accuracy of the measuring device being compared (MSA, 2010). Capability An estimate of the combined variation of measurement errors (random and systematic) based on short-term assessment of the measurement system (MSA, 2010). Cathode Electrode through which electric current flows out of a polarized electrical device. Control Chart A graph of process characterstics, based on sample measurements in time order, used to display the behavior of a process, identify patterns of process variation, assess stability, and indicate process direction (MSA, 2010). Data A collection of observations under a set of conditions that may be variable (a quantified value and unit of measure) or discrete (attribute or counted data such as Pass/Fail or Good/Bad) (MSA, 2010). Drift The actual change in the measurement value when the same characteristic is measured under the same conditions, same operator, at different points in time. Drift indicates how often a measurement needs recalibration (MSA, 2010). Gage Repeatability and Reproducibility An estimate of the combined variation of repeatability and reproducabiilty for a measurement system. The Gage R&R variance is equal to the sum of within-system and between-system variances (MSA, 2010). Linearity The difference in bias errors over the expected operating range of the measurement system. In other terms, linearity expresses the correlation of multiple and independent bias errors over the operating range (MSA, 2010).

14 14 Measurement System A collection of instruments or gages, standards, operations, methods, fixtures, software, personnel, environment, and assumptions used to quantify a unit of measure or fix assessment to the feature characteristic being measured; the complete process used to obtain measurements (MSA, 2010). Measurement System Error The combined variation due to gage bias, repeatability, reproducability, stability and linearity (MSA, 2010). Metrology The science of measurement (MSA, 2010). Out-of-Control State of a process when it exhibits chaotic, assignable, or special cause variation. A process that is out of control is statistacally unstable (MSA, 2010). Part Variation related to measurement systems analysis, part variation represents the expected part-to-part and time-to-time variation for a stable process (MSA, 2010). Performance An estimate of the combined variation of measurement errors based on a long-term assessment of the measurement system; includes all significant and determinable sources of variation over time (MSA, 2010). Pp Process Performance. A simple and straightforward indicator of process performance (MSA, 2010). Ppk Process Performance Index. Adjustment of Pp for the effect of non-centered distribution (MSA, 2010). Precision The net effect of discrimination, sensitivity and repeatability over the operating range of the measurement system (MSA, 2010). Process Control Operational state when the purpose of measurement and decision criteria applies to the real-time production to assess process stability and the measurement or

15 15 feature to the natural process variation; the measurement result indicated the process is either stable or in-control or out-of-control (MSA, 2010). Process Capability compares the output of an in-control process to the specification limits by using capability indices (MSA, 2010) Repeatability The common cause, random variation resulting from successive trials under defined conditions of measurement. Often referred to as equipment variation, although this is misleading. The best term for repeatability is within-system variation when the conditions of the measurement are fixed and defined fixed part, instrument, standard, method, operator, environment, and assumptions. (MSA, 2010) Reproducibility - The variation in the average of measurements caused by a normal condition of change in the measurement process. Typically, it has been defined as the variation in average measurements of the same part between different appraisers using the same measuring instrument and method in a stable environment. This is often true for manual instruments influenced by the skill of the operator. It is not true, however, for measurement processes where the operator is not a major source of variation. For this reason, reproducibility is referred to as the average variation between-systems or between-conditions of measurement (MSA, 2010). Resolution The capability of the measurement system to detect and faithfully indicate even small changes of the measured characteristic. Sensitivity Smallest input signal that results in a detectable output signal for a measurement device. Specification explicit set of requirements to be satisfied (Benbow, 2002). Stability Measurement stability addresses the necessary conformance to the measurement standard or reference over the operating life of the measurement system.

16 16 Tolerance Allowable deviation from a standard or nominal value that maintains fit, form, and fuction (MSA, 2010). Limitations of the Study In this study, the measurement devices tested were limited to two devices that are capable of meeting the JB3 material thickness specification. There may be more accurate measurement devices on the market but the ones chosen to be tested in this study are industry proven devices. The measurement devices also had to be economical because at the end of the study, Ullegheny is to purchase and implement the optimal device and price is a concern. Also, the measurement devices that were selected to be tested didn t need any abnormal environmental controls. Methodology This study is to evaluate two different metrology devices to accurately measure the JB3 titanium cathode material per Company XYZ s design specifications. The devices will be analyzed with their measurement error characteristics, the cost, ease of use, and calibration requirements. The quantitative analysis consists of performing a Gage Repeatability and Reproducibility study, process capability analysis, and range deviation analysis of a standard material.

17 17 Chapter II: Literature Review Continuous improvement is one of the core parts of quality in manufacturing. When it comes to metrology engineering it is a continuous battle to limit the gauge variation, which will in turn reveal a more accurate manufacturing process capability. Quality Costs The benefits of quality are endless and some of the main drivers are cost savings, throughput efficiency, yield savings, and customer satisfaction and confidence. To achieve the most effective quality improvements, management needs to ensure the organization has an understanding of the importance of quality and its benefits. Quality cost reports can be used to point out the strengths and weaknesses in a quality organization. With this information identified, it can be used as leverage to make improvements across an organization. Any reduction in quality costs will have a direct impact on the bottom line margins and is an important issue to an organization (Benbow, 2002). Quality costs are a measure of the costs specifically associated with the achievement or non-achievement of a product or service. Quality costs are broken down into four different categories: Prevention costs are the costs of all activities specifically designed to prevent poor quality in products or services Appraisal costs are the costs associated with measuring, evaluating, or auditing products or services to assure conformance to quality standards and performance requirements. Failure costs are those costs resulting from products or services not conforming to requirements or customer needs

18 18 Total quality costs are the sum of these costs: prevention, plus appraisal, plus failure. It represents the different between the actual cost of a product or service and what the reduced cost would be if there weren t any failures of product or quality defects. (Benbow, 2002) Measurement Devices The measurement device selection is the most important aspect of measurement system analysis. The quality of the measurement system is based on the statistical properties of the data it produces over time (MSA, 2010). Identifying the sensitivity or accuracy of the device is important and there is a common practice to determine this requirement. The commonly known Rule of Tens states that measurement instrument discrimination should divide the total tolerance or process variation into ten parts or more (MSA, 2010). Common industry standards to measure sheet metal are hand tools such as calipers, digital micrometers, and drop indicators. In each of the hand tool categories they range from analog to highly sophisticated digital pressure sensing devices. Depending on the device type, some of these hand tools can be very accurate, however the inspector can introduce a lot of variation due to the manual measurement technique. For instance, a digital micrometer is screw driven and depending on how tight the screw is applied to a thin metal material can significantly change the readout value. Most hand tool measurement devices are relatively cheap compared to other fixed metrology devices. Hand tool measurements are usually very simple to use and commonly found in the sheet metal manufacturing industry. Hand tools, such as the one depicted in Figure 1, are used to give a quick reference to the material thickness but are known not to be accurate enough for final verification.

19 19 Figure 1. Digital hand micrometer from Company XYZ s Corlab Coordinate Measuring Machines (CMM) are very accurate vision or touch probe metrology machines. Usually CMMs are either user controlled or computer controlled systems. Measurements are taken as individual points by a touch probe, optically, or with a laser. They can be programmed manually, then operated in DCC (Direct Computer Controlled) mode which is very repeatable and takes out the human error of interaction. There is little human interaction besides loading the part onto a fixture or location on the machine. Once the part is loaded on the CMM an automated program measures the part. Although a CMM, such as the one pictured in Figure 2, would be a great way to measure the material thickness of sheet metal, they are not commonly used because of cost and programming complexities. Also a measurement system

20 20 like this would be overkill because there are other measurement systems which are much smaller and require less stringent environmental requirements. Figure 2. Direct Computer Controlled Coordinate Measuring Machine The industry standard for measuring the thin metal foils is precision height gauges. These machines are very similar to a digital height indicators but are more accurate and repeatable. Some of these devices are manually operated, where some have electronic actuators to regulate the pressure applied to a material. Most systems come complete with a granite base, plunger actuator, and a digital readout display. Some systems have a plunger that meets a granite base and some have two contacts that read the material thickness between. Depending on the application and measurement, the different contact methods need to be evaluated. The precision

21 21 height gauges are more expensive than the typical hand tool, but less expensive than a CMM. Like hand tools, the precision height gauges like that shown in Figure 3 are very easy to operate and have less stringent temperature and humidity requirements. Figure 3. Heidenhain CT6001 precision height measurement system at Ullegheny Measurement Process A measurement process is a repeated application of a test method with a measurement system. A robust test apparatus and well defined work instruction are essential. A measuring system should be able to provide accuracy capabilities that will assure the reliability of a measurement. (MSA, 2010)

22 22 Readout Readouts consist of indicators, digital readout, and recordings to display the measurement value. Adequate resolution is the degree to which small increments of the measured quantity can be discriminated in the instrument output. This is one very important element to evaluate when trying to identify a measurement system. A typical rule is one digit greater than the least significant digit of the specification. Error in Measurement The difference between the indicated value and the actual value of a measurement quality is error in measurement. Systematic errors are those not usually detected by repetition of the measurement system. It is very important to understand all known sources of error in a measurement system. The requirement of precision measuring devices is that it should be able to represent, as accurately as possible, the dimension it measures. There may be small measurement error but that is why the 10:1 rule is highly recommended when selecting a measurement device (Benbow, 2002). Accuracy Accuracy is the degree of agreement about individual or group measurements with an accepted reference or master value. Measurement science encompasses two basic approaches for determining conformity to measurement accuracy objectives: (1) an engineering analysis to determine all causes of error; (2) a statistical evaluation of data after stripping or eliminating the errors revealed by the engineering analysis (Benbow, 2002, p. 144).

23 23 Precision Precision is the degree of mutual agreement between an individual measurement made under prescribed conditions or how well identically performed measurement agrees with each other. This concept applies to a process or a set of measurements, not to a single measurement, because in any set of measurements, the individual results will scatter about the mean. Since the means of the results from groups of measurement tend to scatter less about the overall mean the individual results, reference is commonly made to the precision of a single measurement as contrasted with the precision of groups of measurements, but this is a misuse of the term. What is really meant is the precision of a set of single measurements or the precision of a set of groups of measurements. (Benbow, 2002, p. 185) Consistency Consistency of the rading on the instrument scale when the same dimension is being measured is necessary. This can easily be tested with any measurement device by making sure the device is at its zeroed state. Then move the scales to its maximum extent and return it back to the zero location. This may be repeated as needed but each time the device is returned to it s zero location the readout should read excactly the same each time (Benbow, 2002). Measurement Variation Measurement system analysis is one overlooked characteristic in many organizations. Assumptions that a system is capable of measuring a certain feature can lead to inaccurate analysis and conclusions when making data driven decisions. When inspectors measure a part inconsistently they may be rejecting good parts and accepting bad parts which are a quality and

24 24 business risk. Inadequate measurement system performance can make the process capability analysis less satisfactory because of the measurement variation induced, shown in Figure 4. Measurement system analysis assesses the statistical properties of repeatability, reproducibility, bias, stability, and linearity. Collectively, these techniques are sometimes referred to as Gage R&R (repeatability and reproducibility) (Breyfogle, 1999, p. 205). + = Product/Process Variation Measurement Variation Observed Variation Figure 4. The relationship between total, process and measuring variation (MSA, 2010) Bias, shown in Figure 5, is the difference between the true value and the observed average of measurements on the same characteristic on the same part. It is a measure of the systematic error of the measurement system and is the main concept of a Gage R&R study. The contribution to the total error comprised of the combined effects of all sources of variation, known or unknown, contributes to the total error and tends to offset consistently. Predictably all results of repeated applications are of the same measurement process at the time of the measurements. (MSA, 2010, p. 206) Possible causes for induced bias are: Instrument needs calibration Worn instrument, equipment, or fixture Worn or damaged master gage can lead to error in master gage

25 25 Improper calibration or use of the setting master Poor quality instrument due to inadequate design Linearity error Wrong gage for the application Different measurement method such as setup, loading, clamping, technique Measuring the wrong characteristic Distortion (gage or part) Environment concerns such as temperature, humidity, vibration, cleanliness Violation of an assumption, error in an applied constant Application such as part size, position, operator skill, fatigue, observation error (MSA, 2010) Figure 5. Measurement bias (MSA, 2010) Stability, or drift, depicted in Figure 6, is the total variation in the measurements observed with a measurement system on the same parts when measured over an extended period of time (MSA, 2010). Stability is one of the areas of concern when

26 26 measuring a feature in manufacturing. If the measurement system was drifting not the process it can create a lot of confusion to solve the drifting process which was really the measurement system. Possible causes for induced bias are: Instrument needs calibration, reduce the calibration interval Worn instrument, equipment or fixture Normal aging or obsolescence Poor maintenance air, power, hydraulic, filters, corrosion, rust, cleanliness Distortion Figure 6. Measurement Stability (MSA, 2010) Linearity, as shown in Figure 7, is the difference of bias through the expected operating range of the equipment. Linearity can be thought of as a change of bias with respect to size (MSA, 2010). The measurement system may be accurate at measuring a small range but its important to test the linearity by measuring the device at the maximum and minimum ranges of the feature you re trying to analyze.

27 27 Possible causes for linearity error include: Instruments that need calibration Worn instrument, equipment or fixture Worn or damaged master gage or error in master gage Distortion changes with the part size (MSA, 2010) Figure 7. Measurement linearity (MSA, 2010) Repeatability, demonstrated in Figure 8, refers to the variability within the appraiser. It is also the variation in measurements obtained with one measurement instrument when used several times by one appraiser while measuring an identical characteristic on the part. The repeatability is within the system variability and this is analyzed with a fixed part and appraiser (MSA, 2010). Possible causes for poor repeatability include: Within-part: form, position, surface finish, taper, sample consistency Within-instrument: repair; wear, equipment or fixture failure, poor quality or maintenance

28 28 Within-standard: quality, class, wear Within-method: variation in setup, technique, zeroing, holding Within-appraiser: technique, position, lack of experience, manipulation skill or training, feel, fatigue Within-environment: short-cycle, fluctuations in temperature, humidity, vibration, lighting, cleanliness Wrong gage for the application (MSA, 2010) Figure 8. Measurement repeatability (MSA, 2010) Reproducibility is referred to as the variability between the appraisers, pictured in Figure 9. Reproducibility is typically defined as the variation in the average of the measurements made by different appraisers using the same measurement instrument when measuring the identical feature on the same part (MSA, 2010). For automated measurement systems often the operator is the main source of variation. Potential sources of reproducibility error include: Between parts: average difference when measuring types of parts A, B, C, etc., using the same instrument, operators, and method.

29 29 Between instruments: average difference using instruments A, B, C, etc., for the same parts, operators, and environment. Between standards: average influence of different setting standards in the measurement process. Between methods: average difference caused by changing point densities, manual versus automated systems, zeroing, holding or clamping methods. Between appraisers: average difference between appraisers A, B, C, etc., caused by training, technique, skill and experience. Between environment: average difference in measurements over time 1, 2, 3 etc., caused by environmental cycles (MSA, 2010) Figure 9. Measurement reproducibility (MSA, 2010) Concept of Gage R&R Study A Gage Repeatability and Reproducibility study is an estimate of the combined variation of repeatability and reproducibility. This in another way is the variance equal to the sum of within system and between system variances. A Gage R&R test can be performed to identify the root cause of the problem in a process, and a breakdown of that data can be seen in an example

30 30 in Figure 10. Measurement system variation can be described by location and width or spread variation (Benbow, 2002). Overall Variation I I I Part-to-Part Variation I Variation due to Gage REPEATABILITY Measurement System Variation I I I I I Variation due to Operators REPRODUCIBILITY I Operator I Operator by Part Figure 10. Breakdown of overall variation (MSA, 2010) The total variation for the Gage R&R study is calculated by summing the square of both the repeatability and reproducibility (R&R) variation and the part-to-part variation and taking the square root. The Gage R&R formula used across the industry is the ANOVA method, shown in Figure 11. This method is a standard statistical technique and it can be used to analyze measurement error and other sources of variability of data in a measurement system. The analysis can be broken down into four categories parts, appraisers, interaction between appraisers, and replication error due to the gage (MSA, 2010) * MS % Tolerance 100 USL LSL

31 31 Figure 11. This figure is showing the percent tolerance calculation. Note the 5.15 standard deviations accounts for 99% of measurement system variation (MSA, 2010) Gage R&R studies are very common in the manufacturing industry to prove whether a measurement system is capable of measuring a specification. Typically Gage R&R studies are performed in the early development of a component to verify they pass the specified requirements. If the Gage R&R passes the requirements, it helps prove that the parts measured are more representative of the manufacturing process and not due to measurement variation introduced which could skew the values. Process Capability, Control Charts, and Statistical Tools The process capability or performance study is how a process is assessed in respect of the specifications. Process capability is analyzed a few different ways and is very sensitive to the input value for the standard deviation. Also, depending on how the data was collected, it needs to be analyzed in a certain manner. There are ways of analyzing a process to itself, and also the process to a given specification. Pp and Ppk outputs, shown in Figure 12, are medical device industry standards for analyzing a process. The Pp is the process capability and the Ppk is the process capability with respect to how centered the process is to the tolerance specification. Figure 12. Pp and Ppk formulas. σ = stdev(x i ) (MSA, 2010)

32 32 Process capability is a way of analyzing a process to determine whether it is in control. Instead of measuring each part manufactured with a statistical sample and analyzing the Pp and Ppk, the entire lot can be accurately depicted. Today there are many statistical tools to analyze the process capability such as software programs like QC Cal and Minitab. These tools are used to make data driven decisions so internal validation testing can be performed. These tools are widely used in the manufacturing industry today to determine the process capability, which is very beneficial to the supplier and customer. The supplier can benefit from seeing a process shift and fixing or adjusting the process before the product goes out of control. If the product goes out of control there will be scrapped material and time wasted. The supplier can also better predict machine wear and preventative maintenance.

33 33 Chapter III: Methodology The problem is the industry standard measurement devices for measuring material thicknesses at sheet metal manufactures do not have the accuracy to measure extremely thin, high surface finish, and tight tolerance metals. There will be two systems evaluated that are benchmarked to the current Vollmer measurement system. First, a Gage R&R will be performed on each system to prove that it s capable of holding the necessary tolerance. Second, a capability analysis will be performed to show the measurement shift and to see the difference between the systems. Finally, a quantitative analysis about the measurement system s characteristics based on cost, ease of use, and calibration requirements. These tests will help identify the measurement system that is needed to properly measure the JB3 titanium cathode material at Ullegheny. Measurement Device Selection and Descriptions Vollmer VMF1000. The Vollmer VMF1000, which is the current thickness measurement device at Ullegheny, has been in service over fifteen years. Even though the device is relatively old, it still is an accurate measurement system. According to Vollmer, the accuracy of the VMF1000 is which is accurate enough to measure the JB3 titanium cathode thickness. (America, 2002) As stated earlier in the paper, this device is not optimal because of the two.500 spheres it has for contact points to measure the material thickness. These contact spheres are not able to be replaced with flat anvil contacts which are required to accurately measure the overall material thickness. Measuring the foil thickness is relatively easy with the Vollmer system pictured in Figure 13. The two contact spheres, shown in Figure 14, have a constant spring load to mate them. A supplied tool is used to spread the contact points. Once the contact points are spread, a foil

34 34 sample can be inserted between the points and the contact points can be gently lowered to the surface of the material. If the contact points are slammed onto the surface it can leave a dent and also give false readings. As long as the contact points are lowered slowly onto the material surface, the Vollmer gives very repeatable results. The measurement taken is then shown on a digital readout display which has a resolution to (America, 2002). Figure 13. Vollmer VMF1000 measurement system at Ullegheny

35 35 Figure 14. Notice the Vollmer VMF1000 sphere contact points circled in the image Fowler THV. The Fowler THV shown in Figure 15 is a horizontal precision measurement system that has two flat anvil.250 in diameter posts, shown in Figure 16, that contact the material to measure the overall thickness. This is the current thickness measuring device at Company XYZ s Corlab. Company XYZ chose this device for calibrating gauge blocks, gauge pins, and measuring precision lengths. The THV is a very versatile measuring device that is capable of measuring the JB3 titanium cathode thickness with an accuracy rating of Measuring the material thickness with the THV can be somewhat complicated because of the setup requirement. Since the contact points can be switched out to measure different

36 36 characteristics, the system is very versatile. A setscrew on each of the anvils holds the posts in place, and it is essential that the setscrew is tight so posts do not move while measuring. When measuring the JB3 titanium cathode the.250 diameter flat anvil posts must be used. Since there are two contact surfaces meeting each other, the calibration of how the two surfaces meet up is very critical to the accuracy of the machine. The calibration company inscribes lines on the posts which need to perfectly line up when setting up the machine. Once the machine is set up, the force of the contact points can be adjusted. It has been determined that ten foot pounds of force is necessary. To open and shut the contact surfaces a round wheel is simply turned and a piece of material can be inserted between the surfaces. The measurement taken is then shown on a digital readout display which has a resolution to (Fowler, 2011).

37 37 Figure 15. Fowler THV measurement system at Company XYZ s Corlab Figure 16. The Fowler THV s two.250 diameter flat anvil contact surfaces Heidenhain CT600. The Heidenhain CT6001 pictured in Figure 17 is a precision length gauge that uses a plunger actuator in the vertical direction that contacts a granite surface, shown in Figure 18. This system is very similar to the THV but instead of being horizontal it is a vertical measurement system. The Heidenhain CT6001 was chosen to be in this study because it was recommended by Company XYZ s calibration company as an alternative to the THV system. The Heidenhain can also be used for calibrating gauge blocks, gauge pins, and measuring precision heights. The Heidenhain CT6001 has an accuracy of which is very capable of measuring the JB3 titanium material thickness specification. The Heidenhain has the versatility of changing out the contact points but it is very simple because the anvils are threaded. When measuring the JB3 titanium cathode material thickness

38 38 the.250 flat anvil must be utilized. The controller is used to raise and lower the plunger actuator and it needs to be set at the ten foot pounds of force mode. To measure the foil, simply place the foil on the granite surface and using the controller lower the anvil until it contacts the material. The measurement taken is then shown on a digital readout display which has a resolution to (Heidenhain, 2011). Figure 17. The Heidenhain CT6001 measurement system at Ullegheny

39 39 Figure 18. The Heidenhain.250 diameter flat anvil utilized to measure the JB3 titanium cathode material thickness. Titanium Cathode Subject Selection and Description The titanium foil samples that were selected for the Gage R&R and the capability study were of grade two titanium of different thicknesses. The ten Gage R&R samples needed to represent normal process variation and also have passing and failing parts. The samples ranged from.0006 to.0014 in thickness to span the entire tolerance band and outside the tolerance band to ensure the measurement system can accurately depict the thickness. The surface finishes for all the samples were 30 Ra minimum, which is representative of the production JB3 titanium cathode material. The thirty samples for the capability study were the best effort JB3 titanium cathode material produced by Ullegheny. Ullegheny cannot define the manufacturing rolling process until a measurement system is identified and implemented. Once the manufacturing rolling process is set, Ullegheny will be able to produce production equivalent material. All of

40 40 the titanium cathode samples will be of the production width which is and cut to a length of Gage Reproducibility and Repeatability Test Description The gage reproducibility and repeatability test was performed per Company XYZ s internal procedure, which complied with ISO and the FDA regulations. The Gage R&R test was performed on the Heidenhain CT6001, Vollmer VMF1000, and the Fowler THV measurement systems. There were ten separate titanium cathode strips, measured by three operators (A, B, and C). Each operator measured each sample three separate times in a blind study. A blind sequence means the operator does know what the part number being sampled but the instructor viewing the Gage R&R does. The operators were trained at the same time with the same procedure. Capability Study Description The capability study was performed after the chosen measurement system was implemented at Ullegheny to show the difference in measurement shift with variation induced. Thirty samples were measured to show the process capability of Ullegheny s Z-Mill manufacturing equipment.

41 41 Gage R&R and Capability Study Data Analysis For both the Gage R&R test and the capability study, MINITAB Version 15 statistical software will be used to evaluate the data. Minitab is a statistical analysis software that has been validated by Company XYZ. The analysis of this software will help make the recommendations for the measurement system through data driven decisions. Gage R&R Data Analysis. Per Company XYZ s variable Gage R&R procedure, the raw data measured by the three operators will be analyzed in Minitab. The Gage R&R (ANOVA) Crossed function in the Minitab software is the Company XYZ specified analysis tool. The crossed formula assumes that the master samples can be selected such that each operator can measure multiple parts from each master sample. Once the analysis has been performed, the results will appear in the Minitab session window. The value that needs to be reviewed is the Total Gage R&R percent tolerance (Study Variation/Tolerance). This value per Company XYZ s standard procedure needs to be lower than 30%, and the lower the value the less measurement variation will be induced. Refer to Figure 11 for the Gage R&R percent tolerance formula to be used. Capability Study Data Analysis. The process capability analysis will be calculated in Minitab using the Process Capability Sixpack function. This function looks at the raw data provided and the upper specification and lower specification limits. The output results to be analyzed are the Pp value and the normality per Company XYZ s procedure. Per Company XYZ s process capability procedure, the Pp needs to exceed 1.00 and the Anderson Darling normality needs to be above.05. If the normality is under.05 it means the data set isn t a normal distribution. Refer to Figure 12 for the formula to calculate Pp. Pp is being analyzed and not Ppk, because Ppk is the process capability with respect to the specification limits. Since one

42 42 manufacturing lot will be evaluated with the measurement systems, there would be bias if a Ppk was evaluated. The Pp analyzes the process capability with respect to the tolerance range. If Ullegheny can hold a tight distribution, it will be simple to move the mean to the specification. Also being evaluated is the range deviation to show how much deviation is present in the measurement systems over fifteen samples. For the capability study evaluation, fifteen samples from one manufacturing lot will be evaluated, and the measurements for all three systems will be taken from the same location on the titanium coupon to eliminate variation between systems. Table 1 Statistical Analysis Studies Gage R&R and Capability Study Total Gage R&R Value Needs to be under 30% Ranking Capability Study Range Deviation - Inches Ranking Capability Study Ppk Value - Normality Pass/Fail Ranking Overall Ranking for Statistical Testing Vollmer Fowler THV Heidenhain Quantitative Analysis of the Measurement System A quantitative analysis is necessary for this study and is based on the measurement system cost, ease of use, setup requirements, and calibration requirement. Ullegheny is very sensitive to the measurement system cost because they have already invested in the Vollmer, so implementing another device needs to be cost effective. In addition, the measurement system that will be utilized will need to be as easy as the Vollmer or easier because of time constraints, and it cannot require a special technician to operate the device. The setup of the Vollmer system

43 43 is very simple so the new system implemented will need to have a fast setup. The calibration requirements for the next system need to have a semi-annual or annual expiration, which would correlate with the rest of Ullegheny s measurement equipment. Table 2 Quantitative Analysis of the Measurement System Measurement System Cost Ranking Setup Requirements Rating 1-10 (1 being the easiest) Ranking Inspection Difficulty Rating 1-10 (1 being the easiest) Ranking Calibration Requirements Rating 1-10 (1 being the same as Ullegheny's current system, 10 being on a different calibration schedule) Ranking Overall Ranking for the quantitative analysis Vollmer Fowler THV Heidenhain Limitations The measurement systems evaluated are limited to two systems besides the Vollmer currently being utilized by Ullegheny. In an ideal world, hundreds of systems could be evaluated but it is not practical for this study. The two best systems that met the cost and accuracy requirements were chosen. The samples being measured by the operators are under strict conditions of not using bias or coaching to drive the results. The limitations of the statistical analysis were per Company XYZ s procedure and only the Minitab software was utilized.

44 44 Chapter IV: Results The purpose of this study was to identify a measurement system to accurately measure the next generation JB3 titanium cathode thickness at Ullegheny. Two measurement systems were identified to be evaluated in the study, as well as the current Vollmer which is utilized to measure the current generation JB3 titanium cathode material thickness. A statistical analysis consisting of a Gage R&R, capability study, and a capability range deviation analysis was performed. In conjunction with the statistical analysis, a quantitative analysis was performed on each of the systems to see how the different key aspects compare to each other. Statistical Analysis: Gage R&R and Capability Study The Gage R&R and capability studies were performed on the Heidenhain, Vollmer, and Fowler THV measurement devices with the trained individuals. The two systems that were to be evaluated for implementation at Ullegheny were the Fowler THV and the Heidenhain systems. The Vollmer measurement device was included to benchmark against the two systems to be evaluated for implementation. Both systems passed the Gage R&R test with total percent tolerances below thirty, but the THV was very close to the specification limits. The Heidenhain had better reproducibility and repeatability values than the THV, but only by four percent, which in percent tolerance variation is a very close deviation. The Gage R&R results were higher than anticipated.

45 45 Table 3 Statistical Analysis Studies Gage R&R and Capability Study Results Measurement System Cost Ranking Setup Requirements Rating 1-10 (1 being the easiest) Ranking Inspection Difficulty Rating 1-10 (1 being the easiest) Ranking Calibration Requirements Rating 1-10 (1 being the same as Ullegheny's current system, 10 being on a different calibration schedule) Ranking Overall Ranking for the quantitative analysis Vollmer $33, Fowler THV $22, Heidenhain $12, The capability study measurements exemplified in Figure 19 were taken from fifteen samples from one manufacturing lot. The two systems evaluated and the benchmark measurement system measured from the same location to eliminate variation and to see the difference in the range deviation between the systems, shown in the graph in Figure 20. The Heidenhain had the least range deviation with.00003, with the Vollmer, and finally with the THV. The range deviations proved to show a high capability with all the systems but once again, the measurements were taken from one manufacturing lot so there was no thickness or surface roughness variation introduced. All three systems surpassed the Pp requirement of 1.0 with the systems having extremely high values. When Pp values are in the range of for all three systems it is showing an extremely high process capability without respect to the specification mean value.

46 46 Figure 19. Capability study for the three measurement devices evaluated Figure 20. Capability study deviation chart for each of the measurement systems.

47 47 The statistical analysis testing proved the Heidenhain and THV systems were acceptable systems to implement at Ullegheny by passing Company XYZ s requirements. The Heidenhain had virtually the same results than the THV measurement system for the capability range analysis with and respectively. The Vollmer system failed the Gage R&R with 37.45% total variation and thus the reason for implementing a more accurate measurement device at Ullegheny. Quantitative Analysis of the Measurement System Results A quantitative analysis was performed on each of the systems based on the measurement system cost, setup requirements, inspection difficulty, and calibration requirements. Since Ullegheny already had the Vollmer measurement system in service, which was a costly device to begin with and has semi-annual calibrations, another measurement system cannot be overburdening. Table 4 Quantitative Analysis Results of the Measurement Systems Measurement System Cost Ranking Setup Requirements Rating 1-10 (1 being the easiest) Ranking Inspection Difficulty Rating 1-10 (1 being the easiest) Ranking Calibration Requirements Rating 1-10 (1 being the same as Ullegheny's current system, 10 being on a different calibration schedule) Ranking Overall Ranking for the quantitative analysis Vollmer $33, Fowler THV $22, Heidenhain $12,

48 48 Measurement System Cost. The benchmark for the cost comparison was the Vollmer VMF 1000 measurement system at Ullegheny. The Vollmer VMF 1000 system is $43,000 and comes with a digital readout and the measurement device, which are separate from one another. The Vollmer system is rather expensive compared to the THV and Heidenhain systems, but in the precision measurement industry it s priced fairly. Even though Ullegheny has had this system in operation for roughly fifteen years, Vollmer still makes this system and it is readily available (America, 2002). The Fowler THV system retails for $22,600 and comes with a Heidenhain digital readout display and the measurement system. The machine is very versatile so different contact tips are also included in the price of the system. The specified contact surfaces, which are the dual.250 diameter flat anvil posts were supplied with the measurement system for the quoted price. The THV system lead time is roughly six to eight weeks because the manufacture builds them to suit the customer needs (Fowler, 2011). The Heidenhain CT6001 measurement system is the cheapest and most accurate out of the three systems evaluated. The complete system cost is $11,500 which includes the digital readout, CT6001 length gauge, granite table with post, switch box, and anvil tip. The price of this system is rather a bargain in the metrology industry but the Heidenhain is limited to only measuring lengths, compared to the THV which can measure an assortment of features (Heidenhain, 2011).

49 49 Setup Requirements. The setup requirements were based upon starting the machine up from its powered off state to being ready to measure samples of the JB3 titanium cathode material thickness. This is a very important aspect because each time Ullegheny manufactures the JB3 titanium; a quality technician will need to prepare the measurement device. The Vollmer system was by far the easiest of the three systems to get from its powered down state to being ready to measure. Simply turn the machine on and take a lint free cloth and slide it under the contact points. Once the contact points are clear of any debris the machine can be set to zero and it is ready to measure. The Vollmer system is very popular with the quality and manufacturing technicians as it is easy to operate. Ranking second out of the three machines was the Fowler THV system. The THV setup was pretty straight forward from turning on the digital readout, inserting the.250 cylindrical posts, cleaning the anvils, and then zeroing the machine. Since the anvil contact posts are removable in the THV the flatness between the contact surfaces are extremely important. In the calibration process lines are inscribed on the anvil posts because they need to be lined up when setting up the machine. In the setup process, when the posts are inserted into the machine it is critical the lines match up or the contact surfaces won t be flat to each other as the calibration company intended. A setscrew holds the anvil posts into the machine, and if the post is not fully seated or the screws are not tight the anvil posts can slip while taking measurements. The setup process can be somewhat tedious to make sure everything is lined up and tightened to ensure the system will be accurate. The Heidenhain setup process is very similar to the other systems evaluated in this study. The digital readout needs to be powered on, granite and.250 diameter anvil surface cleaned, and the plunger actuator lowered to the granite surface to be set at zero. The.250 diameter

50 50 anvil is threaded into the plunger actuator post so it is important this is tightened before setting the machine to zero. Inspection Difficulty. The Vollmer system was very easy to use by simply spreading the contact points and inserting the titanium cathode material. The biggest complaint for the Vollmer system was if the user would open the contact points and release them too fast, the points would slam together and give a false readout. Since the Vollmer has contact points and not surfaces like the other two systems, when the points are slammed they penetrate into the material. It is very apparent when a user has slammed the anvils into the material because the value can be off significantly. When operating this system, if the user is gentle and measuring with care, the false readout issue isn t systemic. This system is able to take measurements very quick with high confidence. It is recommended to clean the contact points if the machine doesn t come back to the zero location after measuring a few samples. The THV system proved to be almost as simple as the Vollmer to use. To measure a sample the round wheel on the THV is turned to open the anvils, and then the wheel can be rotated the opposite direction to close the anvils onto the material. Like the Vollmer system, the THV anvils can be slammed into one another if the user isn t taking care. Since there are two.250 diameter anvils with the THV, the surfaces don t penetrate the titanium material but can compress it slightly. Since it s not as apparent in the readout between a sample that had been slammed or not slammed, this is a concern. The users were instructed to carefully bring the contact surfaces to the titanium and thus the Gage R&R values were acceptable. The THV can measure samples quickly but cleaning the anvils is essential. To clean the anvils a lint free tissue is inserted between the anvils and the contacts surfaces are released to compress the tissue. Then the tissue is to be pulled out. Since the JB43 titanium cathode material is very rough, the peaks

51 51 of the material break off and can stick to the anvils. If the surface particulates build up from not being cleaned it will give false readings. The Heidenhain measurement system was well received by the users in the Gage R&R and capability study. The controller used to raise and lower the plunger actuator and needs to be set at the ten foot pounds of force mode. To measure a sample, place the titanium on the granite surface and, using the controller, lower the anvil until it contacts the material. Just like the THV system, the anvil and granite need to be cleaned after each sample is measured to remove surface particulates on the anvil. Particulate buildup was an issue in preliminary testing but solved with the cleaning step prior to each measurement. The Heidenhain was more sensitive than the THV with the particulate buildup, but since the machine is slightly more accurate, this may be why. Particulate buildup can slow the measurement process up by cleaning the anvil each time, but it is essential for accurate measurements. Calibration Requirements The Heidenhain, Fowler THV, and Vollmer measurement systems are to be calibrated on a semi-annual basis. The calibration company that currently calibrates the Vollmer measurement system at Ullegheny is able to calibrate the Heidenhain and THV system. The calibration is very similar for the three systems so it will not be overburdening to implement the THV or Heidenhain system. Calibrations typically take two-to-three hours depending on how much time the technician needs to spend tweaking the device. Overall Ranking for the Quantitative Analysis The Vollmer was a great benchmarking system to the Fowler THV and Heidenhain systems to baseline what Ullegheny is currently using to what they will be using to measure the JB3 titanium cathode material thickness. In every aspect, the Vollmer was an easier device to

52 52 use, but not as accurate as the THV or Heidenhain systems. The THV and Heidenhain were very similar with the setup and measurement ease of use but the Heidenhain was more sensitive to surface particulate buildup.

53 53 Chapter V: Discussion Company XYZ is an industry leader of producing human implantable defibrillators. The most important components of the defibrillator are the capacitors. The capacitors are charged via the power supply from the battery and then a jolt is delivered through the leads to the heart. The materials that make up the capacitor are the paper insulators, anodes, and cathodes. The next generation defibrillator at Company XYZ needs to be smaller in volume than the current generation but deliver the same amount of energy. The next generation capacitor is called the JB3 and this research paper focuses on the titanium cathode material. Ullegheny is a titanium metal rolling supplier for Company XYZ and makes the current JB3 titanium cathode material. The challenge with the JB3 compared to the JB2 titanium cathode is the roughened titanium material. The overall material thickness of / is the same on both materials but their surface finish requirement is dramatically different. The JB2 has a surface finish of 10 +/- 2 Ra and the JB3 has a 23 Ra minimum. The JB2 material has a very smooth high finish where the JB3 is a rough textured material. Since the JB3 capacitor has less volume than the JB2 material, the added material surface roughness is how Company XYZ is increasing the surface area. The measurement specification required measuring the overall material thickness, and the current JB2 method of measurement did not translate well to the JB3 material. Ullegheny uses a Vollmer measurement system to measure the JB2 titanium cathode material and the anvil tips are spherical shaped. With a very smooth surface finish the spherical shaped anvils can accurately measure the material thickness. When the Vollmer attempted to measure the JB3 titanium cathode, it could not capture the overall material thickness because it measures a point and not a specific area. This prompted the research study to find a measurement device that could measure

54 54 the overall JB3 titanium cathode material thickness and pass the Company XYZ evaluation requirements. The two measurement devices evaluated were the Heidenhain CT6001 and the Fowler THV. These systems were chosen because they are standard measurement devices in the medical device industry. The study was to evaluate both devices and select one to be implemented at Ullegheny to measure the JB3 Company XYZ titanium cathode material thickness. A statistical analysis and quantitative analysis were performed on the system with a comparison to the baseline Vollmer VMF1000, which currently measures the JB2 material thickness. Measurement Device Selection The Heidenhain CT6001 device was chosen to be implemented at Ullegheny to measure the JB3 titanium cathode thickness. The Heidenhain had better statistical evaluation results, which were the most important area to be evaluated in this study. The Gage R&R yielded a total percent tolerance value at 25.5%, where the Company XYZ specification maximum was 30%. The THV s Gage R&R value was extremely close to the 30% limit and this was a concern. Also, the Heidenhain s capability study proved to have less range deviation than the THV. The Pp value from the Heidenhain was a very tight distribution while the THV was not as concentrated. One of the key advantages of the Heidenhain system when compared to the THV is cost. The Heidenhain was over ten thousand dollars cheaper while being more accurate. Although the THV is a more versatile measurement device Ullegheny s plans are to only use the implemented measurement device to measure the JB3 titanium cathode material thickness. The setup and inspection difficulties are very similar between both the systems, with the Heidenhain needing more setup time. Since the Heidenhain device is more accurate, it s understandable that the

55 55 setup and inspection technique can be cumbersome at times. Since the Vollmer system is very easy to use it will be only a slight change for the Ullegheny technicians to start using the Heidenhain measurement system. With the medical device industry pushing for smaller and less invasive devices, quality constraints for suppliers will continue to increase. Suppliers such as Ullegheny will have to accept that it may be more time consuming using the Heidenhain measurement system compared to the Vollmer, but the accuracy results are a priority. With material tolerances tightening with each product generation the statistical quality requirements still need to be met. Limitations In this study, the measurement devices tested were limited to two devices that are capable of meeting the JB3 material thickness specification. There may be more accurate measurement devices on the market but the ones chosen to be tested in this study are industry proven devices. The measurement devices also had to be economical because at the end of the study, Ullegheny was to purchase and implement the optimal device and price is a concern. Also the measurement devices that were selected to be tested didn t need any abnormal environmental controls because Ullegheny s quality lab only meets ISO 9001 requirements. Conclusions This study was very beneficial for Company XYZ because development on the JB3 titanium cathode was halted until a measurement device that could be implemented at Ullegheny to measure the material thickness. Prior to the study, without having an accurate measurement device and procedure to measure the JB3 material thickness, there was no reason to move ahead with the development. Once the Heidenhain measurement system was implemented at

56 56 Ullegheny, the manufacturing and processing improved, to provide JB3 titanium cathode material to Company XYZ.

57 57 References America, V. (2002). VMF brochure. Retrieved from Vollmer America: Benbow, D. W. (2002). The certified quality engineer handbook. Milwaukee: American Society for Quality. Breyfogle, F. W. (1999). Implementing six sigma. Austin, TX: Wiley-Interscience Publication. Fowler. (2011). Fowler/Trimos THV System. Retrieved from Fowler: Heidenhain. (2011). Heidenhain encoders CT6001. Retrieved from Heidenhain Encoders: MSA. (2010). Measurement systems analysis reference manual, 4th Ed.,. Chrysler Corp., Ford Motor Corp., General Motors Corp.,.

58 58 Appendix A: Capability Study Raw and Analysis Data CAPABILITY STUDY OF THE DEVICES Equipment Used: THV Pressure Micrometer Settings:.28 Ft/Lb Applied,.250 Flat Diameter Anvil Location: BSC Qual Lab THV Super Micrometer Accuracy: " Heidenhain Precision Length Gauge Accuracy: " Vollmer Precision Super Micrometer Accuracy:??? Allegheny's Measurement Gauge /-.002" Sample NFowler THV Measurement Data Min Max Avg Range Heidenhain Measurement Data Min Max Avg Range Vollmer Measurement Data Min Max Avg Range

59 59 Individual Value Moving Range Process Capability Sixpack of Vollmer Measurement Data I Chart UCL= ~ =~:~ 1 _ X= LCL= Moving Range Chart 1.>z7t061 UCL= MR= LCL= LSL Capability Histogram USL Normal Prob Plot A D: 0.366, P: Specifications LSL USL I! Jl ID i }---}---} --t-~ - i I I I ---'1"-- t"-. I I I ' Values Last 15 Observations 5 10 Observation 15 Within StDev e-005 C p 3.9 C pk 3.2 Capability Plot Within Overall Specs O v erall StDev e-005 Pp 4.3 Ppk 3.53 C pm * Individual Value Moving Range Process Capability Sixpack of Heidenhain Measurement Data I Chart UCL= ~~~::::=::: 1 _ X= LCL= Moving Range Chart UCL= t;;:;~ : :::: MR= LCL= LSL Capability Histogram USL Normal Prob Plot A D: 0.671, P: Specifications LSL USL I! l id I I I M M I I I I I e I I --r--r--t--,.- -, r --r- 1 I I I I I. -t- - t- - t' - ~ t I I I I.-- I - I +- :--+--L R ft R f Values l ~ ~. ] Last 15 Observations Observation Capability Plot CJ: Within Within StDev e-005 C p 5.01 Overall C pk 3.11 Specs O v erall StDev e-005 Pp 5.95 Ppk 3.69 C pm *

60 60 Individual Value Moving Range Process Capability Sixpack of Fowler THV Measurement Data ~ ('"'?':~ I Chart Moving Range Chart UCL= _ X= LCL= UCL= MR= LCL=0 LSL Capability Histogram USL Specifications LSL USL I! _l id ' I I I --r T Normal Prob Plot A D: 0.711, P: I ~~~ ~----~----r - L.J L. I I " Values L I Last 15 Observations I I I Observation Within StDev e-006 C p C pk 3.3 Capability Plot Within Overall Specs O v erall StDev e-005 Pp 3.58 Ppk 1.18 C pm *

61 61 Appendix B: Gage R&R Raw and Analysis Data Fowler THV Gage R&R Raw Data Part Fowler THV Brian Trial 1 Fowler THV Brian Trial 2 Fowler THV Brian Trial 3 Fowler THV Tom Trial 1 Fowler THV Tom Trial 2 Fowler THV Tom Trial 3 Fowler THV Tony Trial 1 Fowler THV Tony Trial 2 Fowler THV Tony Trial 3 Deviation THV Gage R&R /-.0002" Ti Cathode Thickness Gage name: Date of study : 11/1/2011 Reported by : Tolerance: Misc: Brian Koty k Percent Components of Variation % Contribution % Study Var % Tolerance Results by Parts 0 Gage R&R Repeat Reprod Part-to-Part R Chart by Operators Brian Tom Tony Parts Results by Operators Sample Range Xbar Chart by Operators Brian Tom Tony UCL= _ R= LCL= Brian Tom Operators Operators * Parts Interaction Tony Sample Mean _ X= UCL= LCL= Average Parts Operators Brian Tom Tony Welcome to Minitab, press F1 for help. Gage R&R for Results Gage R&R Study - ANOVA Method Gage R&R for Results

62 62 Gage name: Date of study: 11/1/2011 Reported by: Brian Kotyk Tolerance: Misc: Two-Way ANOVA Table With Interaction Source DF SS MS F P Parts Operators Parts * Operators Repeatability Total Alpha to remove interaction term = 0.25 Two-Way ANOVA Table Without Interaction Source DF SS MS F P Parts Operators Repeatability Total Gage R&R %Contribution Source VarComp (of VarComp) Total Gage R&R Repeatability Reproducibility Operators Part-To-Part Total Variation Process tolerance = Study Var %Study Var %Tolerance Source StdDev (SD) (5.15 * SD) (%SV) (SV/Toler) Total Gage R&R Repeatability Reproducibility Operators Part-To-Part Total Variation Number of Distinct Categories = 16 Gage R&R for Results

63 63 Part Heidenhain Brian Trial 1 Heidenhain Brian Trial 2 Heidenhain Gage R&R Raw Data Heidenhain Brian Trial 3 Heidenhain Tom Trial 1 Heidenhain Tom Trial 2 Heidenhain Tom Trial 3 Heidenhain Tony Trial 1 Heidenhain Tony Trial 2 Heidenhain Tony Trial 3 Deviation Heidenhain Gage R&R Study for /-.002" Ti Cathode Gage name: Date of study : Reported by : Tolerance: Misc: Sample Range Sample Mean Percent Components of Variation Results by Parts 18 % Contribution % Study Var % Tolerance I ~~~ Gage R&R Repeat Reprod Part-to-Part Parts R Chart by Operators Results by Operators Brian Tom Tony j.. = Jl ~ I li: ~I UCL= _ R= LCL=0 Xbar Chart by Operators Brian Tom Tony V[L~~~ I _ X= UCL= LCL= Brian Tom Tony Operators 11:':::1 Operators * Parts Interaction Operators Brian Tom Tony Parts Average Welcome to Minitab, press F1 for help. Retrieving project from file: 'C:\DOCUMENTS AND SETTINGS\G044904\DESKTOP\MEASUREMENT SYSTEM SELECTION\GRR TEST DATA\HEIDENHAIN GRR TEST.MPJ' Gage R&R Study - ANOVA Method Two-Way ANOVA Table With Interaction Source DF SS MS F P Parts Operators Parts * Operators Repeatability Total

64 64 Alpha to remove interaction term = 0.25 Gage R&R %Contribution Source VarComp (of VarComp) Total Gage R&R Repeatability Reproducibility Operators Operators*Parts Part-To-Part Total Variation Process tolerance = Study Var %Study Var %Tolerance Source StdDev (SD) (5.15 * SD) (%SV) (SV/Toler) Total Gage R&R Repeatability Reproducibility Operators Operators*Parts Part-To-Part Total Variation Number of Distinct Categories = 19 Gage R&R for Results Part Vollmer Brian Trial 1 Vollmer Brian Trial 2 Vollmer Gage R&R Raw Data Vollmer Brian Trial 3 Vollmer Tom Trial 1 Vollmer Tom Trial 2 Vollmer Tom Trial 3 Vollmer Tony Trial 1 Vollmer Tony Trial 2 Vollmer Tony Trial 3 Deviation

65 65 Vollmer Gage R&R Study /-.0002" Ti Cathode Thickness Gage name: Date of study : Reported by : Tolerance: Misc: Components of Variation Results by Parts Percent % Contribution % Study Var % Tolerance Sample Range Gage R&R Repeat Reprod Part-to-Part R Chart by Operators Brian Tom Tony Xbar Chart by Operators Brian Tom Tony UCL= _ R= LCL= Parts Results by Operators Brian Tom Operators Operators * Parts Interaction 8 9 Tony 10 Sample Mean _ X= UCL= LCL= Average Operators Brian Tom Tony Parts Welcome to Minitab, press F1 for help. Gage R&R Study - ANOVA Method Two-Way ANOVA Table With Interaction Source DF SS MS F P Parts Operators Parts * Operators Repeatability Total Alpha to remove interaction term = 0.25 Gage R&R %Contribution Source VarComp (of VarComp) Total Gage R&R Repeatability Reproducibility Operators Operators*Parts Part-To-Part Total Variation

66 66 Process tolerance = Study Var %Study Var %Tolerance Source StdDev (SD) (5.15 * SD) (%SV) (SV/Toler) Total Gage R&R Repeatability Reproducibility Operators Operators*Parts Part-To-Part Total Variation Number of Distinct Categories = 12 Gage R&R for Results

67 67 Appendix C: Heidenhain work instructions for Ullegheny Ludlum for measuring the JB3 Company XYZ titanium cathode material thickness. Heidenhain Length Gauge Work Instructions for Ullegheny Ludlum These work instructions are intended for measuring the Company XYZ JB3 titanium cathode material thickness A - Setting up the Heidenhain ID No F 1) Power the Heidenhain ND 287 readout to the On position The power switch is on the back of the unit.

68 68 2) Press any key to get the readout screen to appear 3) With switch box move the length gauge slightly down until the reference mark is crossed. This will produce a live readout 4) Make sure the force is set to 3 which is the maximum Newton s force applied by the Heidenhain

69 5) Clean the System - With lint free wipes and alcohol clean wipe the granite base, flat anvil, and.005 Master block. - Lower the indicator onto a lint free wipe then pull it out. This will help clean any residue off of the surfaces. Make sure no small pieces of lint are trapped between the indicator and granite surface. *** This step is essential for the machine to produce accurate measurements *** 69

70 70 6) Master the machine Lower the indicator until it touches the granite base. Press 0 into the readout and hit enter. Move the indicator up and down a few times to verify it consistently reads out when the anvil is contacting the granite base. *** This is a very important step *** 7) Measure the cleaned.005 gauge block to verify the system has been mastered correctly..005 Master Gauge Block

71 71 B - Measuring the Raw Material 1) Place pristine wrinkle free material under the indicator. Lower the indicator until the digital readout stops descending in value. This will be the measured thickness to record. 2) Before each measurement is taken lower the indicator until it contacts the surface. Verify the readout is at and if it isn t there is debris on the indicator or granite surface. Refer to section A4 to clean the contacts and re-zero the device. This is very important because if you don t verify it s zeroing out before each measurement you will get false readings.

72 C Shutting down the Heidenhain. 1) Raise the indicator until it reaches the upper stop point. (Do not leave the indicator in the lowered position!) 2) Turn power button on the readout display to the off position. 72