1. Matkap uçlarının Ark PVD ile Tin kaplanmasında proses parametrelerinin Taguchi metodları ile optimizasyonu
- Author
-
Keleş, Özgül, Taptık, İsmail Yılmaz, and Diğer
- Subjects
Coating ,Optimization ,Metalurji Mühendisliği ,Tin ,Metallurgical Engineering ,Taguchi method ,Twist bit - Abstract
ÖZET Bu çalışmada Ark PVD yöntemi ile TİN kaplı matkap uçlarının üretiminde müşteri istekleri doğrultusunda belirlenmiş kalite karakteristiklerine ulaşılması için kaplama proses parametrelerinin optimizasyonuna yönelik olarak Taguchi Metodu uygulanmıştır. Taguchi'ye göre kalite ürün müşteriye ulaştıktan sonra ortaya çıkan şikayet ve geri dönüşleri bağlı bir kayıptır. Taguchi kaliteyi `spesifikasyonlara uygunluk` olarak değil `hedef değere yaklaşma` olarak tanımlamıştır. Ona göre ürün kalitesi, belirlenen hedef değere yaklaştıkça artarken kayıp azalmakta, kalite hedef değerden uzaklaştıkça kayıp artmaktadır. Bu amaçla TİN kaplı takımların ülkemizde ve dünyada kullanımının yeniliği ve henüz bu konu ile ilgili olarak hazırlanmış bir standarun eksikliğinden dolayı müşterilerin kaplamalı takımlardan beklentilerini belirlemek amacıyla bir anket çalışması yapılmıştır. Bu anket çalışması sonucunda müşteri dili ile ifade edilmiş ve teknolojik olarak elde edilebilir kalite karakteristikleri QFD yöntemi ile mühendis diline çevrilmiştir. Taguchi Metodu içerisinde yer alan tasarım prosesi sistem tasarımı, parametrelerin tasarımı ve toleransların tasannu olmak üzere üç aşamalı olarak incelenmiştir. Bu çalışmada TİN kaplamalı ürün üretiminde öngörülen hedef değerlerinin ulaşmaya yönelik olarak Ark-PVD prosesinde parametre tasarımı incelenmiştir. Taguchi metodunu diğer deneysel tasarım yöntemlerinden ayıran en önemli fark ön hazırlık aşamasıdır, ön hazırlık aşamasında Ark-PVD prosesine ve belirlenen kalite karakteristiklerine etki eden proses parametrelerinin belirlenmesi için bir takım oluşturulmuştur. Bu takım öncelikle proses akış diyagramım oluşturmuş daha sonra beyin fırtınası yardımıyla Ark-PVD prosesi için sebep-sonuç diyagramı hazırlanmıştır. Bu diyagramda prosese etki eden parametreler çeşitli gruplar altında toplanmış ve bu parametrelerin her biri sebep, ulaşılmak istenen kalite karakteristikleri ise sonuç olarak ele alınmıştır. Bu amaca yönelik TİN kaplı takımların kalite karakteristikleri doğrultusunda Ark-PVD prosesi için sebep-sonuç diyagramı yardımıyla belirlenmiş proses parametreleri yapışma, yüzey pürüzlülüğü, sertlik ve kaplama kalınlığı olmak üzere dörde mdirilmiştir. Son aşamada takım tarafından deneylerde ele alınmak üzere seçilen her bir proses parametresi için üç seviye belirlenmiştir. Seviyelendirilen bu parametreler daha sonra Taguchi metodunda yer alan L934 ortogonal dizinine yerleştirilmiştir. Daha sonra L^'una göre denemeler yapılmıştır. Sonuçta elde edilen TİN kaplı takımlarda kalite karakteristiklerinin belirlenmesi amacıyla değişik ölçümler yapılmıştır. xiii OPTIMIZATION OF PROCESS PARAMETERS OF ARC-PVD TECHNIQUE FOR THE PRODUCTION OF TsN COATED DRILL PINS BY TAGUCHI METHODS SUMMARY The importance of quality concept and quality activities have shown a steady increase for the last couple of years. The first serious meeting of quality with global world market had encountered after the second World War. Especially Japanese and American businessmen, scientists and engineers had consumed a lot of time on quality and quality tools, techniques for producing high quality and low cost products. The changing of the world market caused the changing the meaning of quality. In 1960's quality was defined as `the conformance of specifications` by Juran. Nowadays, because of the customs unions, developing transportation and communication systems, the competition conditions of world market get though.. To increase their world market shares the firms have to provide high quality and low cost which are the basic competition conditions. In the light of production the meaning of quality changed with `customer satisfaction`. In many companies including product based businesses the primary way to increase customer satisfaction is improving customer service rather than by providing products of superior utility or functional value of users. Yet there is a growing acknowledgment that has the product performs, relative to customer's ideals and expectations, is the primary determinant of customer satisfaction. This has led to more emphasis on product design and redesign to provide product that truly satisfies the customer. But companies must know what customer want and they must have quality assurance system to provide this. Producers have to achieve ISO 9000 standards m order to establish quality assurance. ISO 9000 standards were developed by the technical committee of 176 International Standards Organization in 1987. There are five standard series: ISO 9000 Quality management and quality assurance standards guidelines, ISO 9001 contains design and production stages, ISO 9002 is the standards of only production stages, ISO 9003 contains final control and measurements, ISO 9004 contains the most widely used guidelines internationally, xivThe firms which reach design, production and service quality have to provide new form of management. This human motivation based management is called Total Quality Management. (TQM). The superiority of TQM from the others is the use of quality tools and techniques in the research and development activities. Quality tools are used to solve problem and to improve process/ product quality in production, design and service. The basic statistical quality tools which are defined by Ishikawa are; 1. Pareto analysis, 2. Cause and Effect Diagram, 3. Histogram, 4. Cheek Sheet, 5. Stratification, 6. Control Chart and 7. Scatter Diagram Quality techniques are used for improving the product in design. These quality techniques are Quality Function Deployment (QFD), Failure Modes and Effect Analysis (FMEA), Failure Tree Analysis (FTA), Statistical Quality Control (SPC) and Design of Experiments (DoE). In all DoE methods, the first stage is started problem definition. Using QFD method, problem can be defined and at the same time customer requirements are determined correctly. The application of QFD method in design stage customer language is translated to engineer language. Thus, customer-producer and technology chain is built. FMEA, FTA and SPC techniques are employed during the production stages to analyze the problem. By means of FMEA and FTA possible failures and effects which occur in production phase are determined. SPC techniques are used to show the points which are out of control in production phase. Establishing the necessary properties which will be controlled are determined by DoE techniques. Firstly, data must be collected properly and methods should be applied in coordination in order to get the advantages of these techniques on product and production quality, The goal of DoE methods is to search for the causes of variation and to remove them or robust the system under investigation against variations. Before applying these techniques data which are collected by means of other quality techniques are evaluated, parameters are determined and selected. Figure 1 shows the parameter optimization flow chart according to different experimental design techniques. The aim of all DoE methods is to optimize the system under investigation to achieve low cost and high quality. But the difference is vanished in applications. Experimental design techniques according to application types are: xv. The Classical methods,. The Shainin methods and. The Taguchi methods. The Genetic Algoritm (optimization by means of mutation rules). eotuütioas Problem analysis (Cause and Effect, FMEA,FTA) Taguchi experiment design Selecting suitable experiment If necessary, pseudo factor design or multi-level design including interactions 1 Homing-in according to Taguchi Multi-variat» charts Componenj Pairred compansons yksa good/bad leveuK_ >w known? yS Figure 1. Flowchart. Complete design of experiments (Full factorial) Variable search Scatter plots Carriying out experiment XVI. The Classical Methods Classical experimental design techniques include the oldest and also the most difficult methods. These are classified into three groups ; one factor at a time, full factorial and fractional factorial designs. If in one treatment only one parameter is changed and the others hold constant this is called one factor at a time methods. Interaction between parameters can not be determined by this method. In contrast to one factor at a time method full factorial designs and fractional factorial design include interactions. As a result increasing treatment number depends on the parameter number and its levels in all classical experimental design techniques.. The Shainin Methods The difference between Shainin methods and others is shown in problem analysis stages. Pretreatment stage is considerably important part in Shainin technique. In this part to determine and to reduce parameters various special tools - components search, multivariate charts and paired comparison- are used. Therefore, the investigated system must already exists and experimenter must have enough knowledge about it. The Taguchi Methods This experimental design technique studies was accelerated in Japan and Taguchi technique is developed by Dr. Genichi Taguchi. Although Taguchi methods involve the usage of statistical techniques -such as orthogonal arrays and partition of total sum of squares (analysis of variance )- they are not statistical. These properties are criticized by statisticians and mathematicians. Taguchi has given importance in to pretreatment stage in experimental design like Shainin but he has not used any special tools. Some scientists criticize Taguchi techniques, due to the lack of rules for decreasing parameters and interactions. The term Taguchi Methods refers to parameter design, tolerans design, the quality loss function, on line quality control, design of experiments using orthogonal arrays and methodology applied to evaluate measuring systems. Seven points explain the basic elements of Genichi Taguchi's quality philosophy in a nutshell. 1. An important dimension of quality of manufactured product is total loss generated by product to society. 2. In a competitive economy, continuous quality improvement and cost reduction are necessary for staying in business. 3. A continuous quality improvement program includes incessant reduction in variation of product performance characteristics about their target values. xvu4. The customer's loss due to a product's performance variation is often approximately proportional to the square of deviation of performance characteristic from its target value. 5. The final quality and cost of a manufactured product are determined to a large extent by engineering designs of products and its manufacturing process. 6. A product's (or process') performance variation can be reduced by exploiting the non linear effects of the product (or process) parameter on the performance characteristics. 7. Statistically planned experiments can be used to identify the settings of product (and process) parameters that reduce performance variation. The section on Taguchi's philosophy described the three phases of the Taguchi Method-System design, parameter design and tolerance design. In system design experimenter should use scientific and engineering concepts to develop prototype and select appropriate product and process parameters such as materials, machines, tools, etc. In parameter design, experimenter should use engineering and statistical concepts to determine optimum settings of selected parameters. In tolerance design, experimenter should use engineering and statistical concepts to further reduce variability in performance statistics. In this study Taguchi Methods have been used to optimize TiN coated HSS drill pins parameters by means of Arc PVD coating technology. The most important difference between Taguchi and other experimental design methods is the pretreatment stage. In this stage the following activities have been achieved;. Cause and effect diagram has been formed by means of brain storming.. An enquiry had been prepared to learn what is the customer requirements on cutting tools. This activity is a part of QFD techniques. Collected data from enquiry and cause and effects diagram had been discussed by team and quality characteristics are selected. Enquiry results and fishbone diagram help team the selection of quality characteristics and parameters. In fishbone diagram, cause equals parameter and effect equals quality characteristic. Team has determined three levels for each parameters and they are given Table 1. These selected parameter and quality characteristics are replaced by an ortogonal array 0U34) and treatment that is achieved depends on this orthogonal. Following the treatment characteristics are measured by using different measurement systems. xvuiTabic 1. Parameters and its levels. As a result of this study the analysis of each quality characteristics is achieved by means of ANOVA techniques. ANOVA analysis shows the optimum level for quality characteristics. In Table 2 observed quality characteristics values are shown in L9 (34) ortogonal array. Table 2. L9 (34) ortogonal array. Consequently it is determined that the chosen parameters have limited affect on the quality characteristics of TiN coated drill pins. Thus, the TiN coating process can be applied in a relatively wide range. XIX 240
- Published
- 1996