1 DEVELOPMENT OF HIGH SPEED INTERNAL FINISHING AND CLEANING OF FLEXIBLE CAPILLARY TUBES BY MAGNETIC ABRASIVE FINISHING By JUNMO KANG A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FU LFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2012
2 2012 Junmo Kang
3 I dedicate this t o my family
4 ACKNOWLEDGMENTS I would like to thank my f amily for giving me the chance to study in the US and I would like to thank my advisor, Dr. Hitomi Yamaguchi Greenslet, for her great research guidance and total support. Moreover, this work would not have been possible without her and all the members of the MTRC (Machine Tool Research Ce nter). Therefore, I really would like to thank them with sincerity. I would also like to thank the Creganna Tactx Medical Company for providing the workpiece s the University of Florida Research Foundation (Gatorade) the Office of Research at the Universi ty of Florida for travel grant s and the 2011 2012 Society of Manufacturing Engineers (SME) Education Foundation (E. Wayne Kay Graduate Scholarship) for their support and encouragement. Special thanks to M r. John Greenslet for his kind assistance with my w ritten work.
5 TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ .. 4 LIST OF TABLES ................................ ................................ ................................ ............ 7 LIST OF FIGURES ................................ ................................ ................................ .......... 8 ABSTRACT ................................ ................................ ................................ ................... 12 CHAPTER 1 INTRODUCTION ................................ ................................ ................................ .... 14 1.1 Background ................................ ................................ ................................ ....... 14 1.1.1 Introduction of Flexible Capillary Tubes and Necessity of Internal Cleaning ................................ ................................ ................................ ........ 14 1.1.2 Current Techniques for Internal Machining of Tube s and Magnetic Abrasive Finishing ................................ ................................ ......................... 16 1.2 Development of Internal Cleaning and Finishing Process for Flexible Capillary Tubes using Magnetic Abrasive Finishing ................................ ............ 19 2 INTERNAL FINISHING PROCESS USING MAGNETIC ABRASIVE FINISHING .. 22 2.1 Magnetic Abrasive Finishing ................................ ................................ ............. 22 2.2 Processing Principle and Parameters of Internal Finishing of Straight Tube .... 25 3 PARAMETERS AFFECTING ON FINSIHNG CHARACTERISTICS IN FLEXIBLE CAPILLARY TUBE FINISHING USING A SINGLE POLE TIP SYSTEM ................................ ................................ ................................ ................. 30 3.1 Processing Principle of Flexible Tube Finishing ................................ ................ 30 3.2 Effects of Finishing Time, Tube Rotation, Slits, a nd Abrasive Size on Edge and Surface Finishing ................................ ................................ .......................... 33 3.2.1 Experimental Setup and Conditions ................................ ........................ 33 3.2.2 Effects of Diamond Abrasive S ize on Deburring Flexible Capillary Tube ................................ ................................ ................................ .............. 38 3.3 Roles and Behaviors of Magnetic Tools ................................ ............................ 46 3.3.1 Experimental Setup and Conditions ................................ ........................ 51 3.3.2 Effects of Material Properties of Magnetic Tools ................................ ..... 52 3.3.3 Hybrid Finishing Methods for Flexible Capillary Tube .............................. 60 4 MULTIPLE POLE TIP SYSTEM USING A METASTABLE AUSTENITIC STAINLESS STEEL TOOL ................................ ................................ ..................... 63 4.1 Introduction of Multiple Pole tip System ................................ ............................ 63
6 4.2 Magnetic Properties of Metastable Austenitic Stainless Steel Tool .................. 65 4.3 Finishing Characteristics in Multiple Pole tip System ................................ ........ 67 4.3.1 A Method to Deliver Magnetic Abrasive Deeper into Capillary Tube ....... 67 4.3.2 Effects of Heat treated Sections on Internal Deburring of Flexible Capillary Tubes with Multiple Laser machined Slits ................................ ...... 72 4.3.3 Effects of Heat treated Sections on Finished Surface and Abrasive Behavior ................................ ................................ ................................ ........ 74 4.3.4 Pole tip Feed Length and Surface Uniformity ................................ .......... 85 5 DEVELOPMENT OF HIGH SPEED FINISHING MACHINE FOR INTERNAL FINSHING OF CAPILLARY TUBES ................................ ................................ ....... 90 5.1 Design and Construction of High speed Finishing Machine .............................. 90 5.1.1 Proposal of New Finishing Machine ................................ ........................ 90 5.1.2 Description of New High speed Finishing Machine ................................ 94 5.2 Finishing Characteristics of High speed Machine ................................ ............. 97 6 HIGH SPEED INTERNAL FINISHING OF CAPILLARY TUBES BY MULTIPLE POLE TIP SYSTEM ................................ ................................ .............................. 102 6.1 Internal Finishing of Capillary Tubes by High speed Multiple Pole tip System 102 6.2 Effects of Lubrication on Finishing Characteristics ................................ .......... 106 6.3 Effects of Tube Rotational Speed on Finishing Characteristics ...................... 108 6.4 Flexible Capillary Tube Finishing ................................ ................................ .... 112 6.4.1 Finishing Characteristics ................................ ................................ ....... 112 6.4.2 Tool Behaviors and Particle Distribution ................................ ................ 119 6.4.3 Relationship between Finishing Force and Magnetic Tools ................... 121 6.4.4 Effe cts of Tube Magnetism on Finishing Characteristics ....................... 125 7 DISCUSSIONS AND CONCLUSIONS ................................ ................................ 130 7.1 Fundamental Finishing Characteristi cs of Flexible Capillary Tubes ................ 130 7.2 A New Finishing Method: Multiple Pole tip Finishing System ......................... 131 7.3 Development of High sp eed Internal Finishing Machine ................................ 132 7.4 Internal Finishing of Capillary Tubes by High speed Multiple Pole tip System: Straight Capillaries ................................ ................................ ............... 133 7.5 Internal Finishing of Capillary Tubes by High speed Multiple Pole tip System: Flexible Capillaries ................................ ................................ ............... 133 8 REMAINING WORK ................................ ................................ ............................. 135 8.1 Development of Higher Numbered Pole tip System ................................ ........ 135 8.2 Pole Rotation System for Flexible Capillary Tubes ................................ ......... 136 8 .2 Magnetic Field Analysis of Multiple Pole tip System ................................ ....... 136 LIST OF REFERENCES ................................ ................................ ............................. 137 BIOGRAPHICAL SKETCH ................................ ................................ .......................... 143
7 LIST OF TABLES Table page 3 1 Experimental conditions for internal deburring of flexible capillary tube with laser machined slits: Tube A ................................ ................................ ............... 36 3 2 Experimental conditions of internal finishing of straight capillary tube ................ 52 3 3 Experimental conditions of internal deburring of flexible capillary tube wi th laser machined slits: Tube B ................................ ................................ ............... 61 4 1 Experimental conditions for 9 8 18 and 18 8 9 pole tip set ................................ 71 4 2 Experimental conditions for 18 18 18 pole tip set ................................ ............... 81 5 1 Speci fications of finishing machines ................................ ................................ ... 92 5 2 Experimental condition s for finishing performance between existing low speed machine and newly developed high speed finishing machine ................. 98 6 1 Experimental conditions for the high speed internal finishing of straight capillary tubes ................................ ................................ ................................ ... 105 6 2 Experimental conditions for the high speed internal finishing of flexible capillary tubes: Tube B ................................ ................................ ..................... 114
8 LIST OF FIGURES Figure page 1 1 Schematic of electropolishing process ................................ ............................... 17 2 1 Schematic of the processing principle for planar magnetic abrasive finishing .... 23 2 2 Schematic of the processing principle for cylindrical magnetic abrasive finishing ................................ ................................ ................................ .............. 24 2 3 Schematic of processing principle of internal finishing of straight tube using rotating tube stationary pole system ................................ ................................ .. 25 2 4 Microscopy and scanning electron microscopy ................................ .................. 27 3 1 Schematic of processing principle of internal finishing of flexible tube ............... 31 3 2 Exterior and interior microscopy of laser ma chined flexible capillary tubes ........ 31 3 3 Photographs of tubes chucked on the finishing spindle with differen t slit patterns ................................ ................................ ................................ .............. 32 3 4 External view of finishing equipment for fle xible capillary tube finishing ............. 34 3 5 Micrographs of unfinished surface, and surfaces finished for 30 and 60 min with 50 70 m diamond abrasive ................................ ................................ ........ 39 3 6 Changes i n material removal with finishing time ................................ ................. 40 3 7 Micrographs of surface finished with 20 40 m diamond abrasive and 4 8 m diamond abrasive ................................ ................................ ............................... 41 3 8 Three dimensional burr shap es measured by optical profiler ............................. 43 3 9 Three dimensional surface sha pes measured by optical profile r ........................ 44 3 10 Micrographs of outer surface of tube; as received surface, surface finished with 4 8 m diamond abrasive, and surface finished with 20 40 m diamond abrasive ................................ ................................ ................................ .............. 45 3 11 Ch anges in material removal with finishing time according to diamond abrasive size ................................ ................................ ................................ ....... 45 3 12 Schematic of iron particle, magnetic abrasive, and diamond abrasive behavior inside the flexible capill ary tube with magnet feed ............................... 47 3 13 Schematics of processing principle with different magnetic tools ....................... 47
9 3 14 Photographs of c arbon steel rod and 304 stainless steel rod inside nonferrous polymer tube ................................ ................................ ..................... 49 3 15 Photographs and surface roughness profiles of tube interior before finishing, surface finished with iron particles, surface finished with carbon steel rod, and surface finished with stainless steel rod ................................ ...................... 53 3 16 Three dimensional surface shapes of tube interior before fi nishing, surface finished with iron particles, surface finished with carbon steel rod, and surface finished with stainless steel rod ................................ ............................. 55 4 1 Schematics of single pole tip system and multiple po le tip system .................... 64 4 2 Tool geometry and photograph of partially heat treated stainless steel tool with iron particles ................................ ................................ ................................ 66 4 3 Part ially heat treated stainless steel tool and X ray diffraction patterns ............. 67 4 4 Geometry of pole tip set geometry and pole tip shapes and changes in m agnetic flux density ................................ ................................ .......................... 69 4 5 External view of multiple pole tip system (9 8 18 pole tip set) ............................ 70 4 6 Changes in surface roughness with distance X ................................ .................. 71 4 7 Intensity maps of surface and three dimensional surface shapes measured by optical profiler ................................ ................................ ................................ 73 4 8 Magnet ized tools A, B, C and D with iron particles ................................ ............. 75 4 9 Motions of each tool with mixed type magnetic abrasive for a pole tip stroke length of 18 mm ................................ ................................ ................................ .. 76 4 10 Geometry of single and multiple pole tip sets ................................ ..................... 79 4 11 Changes in magnetic flux density of single and multiple pole tip system with distance X ................................ ................................ ................................ ........... 81 4 12 External view of multiple pole tip system (18 18 18 pole tip set) ........................ 82 4 13 Changes in surface roughness with distance X ................................ .................. 83 4 14 Changes in pole tip coverage times and surface roughness with distance X ..... 86 4 15 Three dimensional surface shapes measured by optical profiler ........................ 88 5 1 External view of the low speed finishing machine ................................ .............. 91 5 2 External view of existing high speed finishing machine ................................ ...... 93
10 5 3 Schematic of the proposed high speed finishing machine ................................ .. 94 5 4 Photographs of developed machine overview, finishing unit, and control u nit .... 95 5 5 Photograph of finishing unit and pole tip geometry ................................ ............. 96 5 6 Microscopy of tube interiors before and after finishing ................................ ....... 99 5 7 Three dimensional surfaces of tube interiors before and after finishing at the rotational speeds of 2500 min 1 and 30000 min 1 ................................ .............. 100 5 8 S urface roughness of unfinished and finished surfaces in rotational speed 2500 and 30000 min 1 ................................ ................................ ....................... 101 6 1 Schematic of the processing principle for internal finishi ng by multiple pole tip system ................................ ................................ ................................ .............. 103 6 2 Changes in magnetic flux density in multiple pole tip sets at Y =0 mm ............ 104 6 3 Tool geometry and magnetized t ool with iron particles ................................ ..... 106 6 4 Microscopy of the initial and finished surface for 10 min at 30000 min 1 ........... 107 6 5 Intensity maps and oblique plots of the initial surface and the surface finished for 10 min at 5000 min 1 10000 min 1 20000 min 1 30000 min 1 ..................... 110 6 6 Intensity maps and oblique plots of the surface finish ed for 2 0 min at 5000 min 1 10000 min 1 20000 min 1 ................................ ................................ ........ 111 6 7 Changes in surface roughness with tube revolution observed in multiple pole tip system ................................ ................................ ................................ ......... 112 6 8 Changes in material removal with tube revolution in multiple pole tip system .. 112 6 9 Schematics of the processing principle for the internal finishing of flexible capillary tubes by a multiple pole tip system ................................ .................... 113 6 10 Schematics of the partially heat treated metastable austenitic stainless steel tools ................................ ................................ ................................ .................. 113 6 11 External view of high speed multiple pole tip finishing system for flexible capillary tube finishing ................................ ................................ ...................... 114 6 12 Microscopy of tube in teriors: As received condition and s urface finished for 26 min by Tool C and Tool D ................................ ................................ ............. 116 6 13 Spectrum plots of the tube interiors: As received condition, surface finished by T ool C and T oo l D ................................ ................................ ......................... 117
11 6 14 Microscopy of tube exteriors: As received condition and surface finished for 26 min ................................ ................................ ................................ ............... 118 6 15 Powder distribution wi th differently heat treated magnetic tools ....................... 120 6 16 Schematics of the powder distribution with differently heat treated magnetic tools ................................ ................................ ................................ .................. 120 6 17 Schematics of magnetic force measurement system ................................ ....... 121 6 18 Photograph of magnetic force measurement system for multiple pole tip finishing m achine, Tool C with particles, and Tool D with particles ................... 122 6 19 Relationship between magnetic force and ferrous tool ................................ ..... 123 6 20 Relationship between finishing pressure and heat treated tools ....................... 124 6 21 2 D magnetic field analysis by FEM with non ferrous and ferrous tubes .......... 126 6 22 Relationship between magnetic flux density and both capillary tubes at distance X ................................ ................................ ................................ ......... 128
12 Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy DEVELOPMENT OF HIGH SPEED INTERNAL FINISHING AND CLEANING OF FLEXIBLE CAPILLARY TUBES BY MAGNETIC ABRASIVE FINISHING By Junmo Kang December 2012 Chair: Hitomi Yamaguchi Greenslet Ma jor: Mechanical Engineering Tube shaped medical components such as catheter shafts, coronary stents and biopsy needles should be precisely designed and manufactured since surface imperfections can cause severe possibly lethal implications for patients Therefore, i nternal machining (surface and edge finishing) is increasingly in demand for these devices but as the tube diameter decreases, the more difficult the internal finishing becomes M agnetic abrasive finishing (MAF) has been studied to accomplish the internal finishing of capillary tubes The MAF process can produce a smoothly finished inner surface of tubes by means of relative motion between magnetic abrasive and tube inner surface. To achieve successful internal finishing of capillaries control ling the magnetic abrasive motion and lubricant inside the tube is critical. The magnetic abrasive is suspended in magnetic field generated by the magnet located outside the tube. The magnetic field is controlled by the geometry and arrangement of the magn ets as well as the geometry of pole tip attached to the magnets. In the case of flexible capillary tubes, deburring the edges of laser machined slots and removing the re solidified material caused by laser machining process on the inner surface are require d in addition to the
13 internal surface finishing. The burrs are obstacles to magnetic abrasive hindering the generation of smooth relative motion against the tube inner surface. The tube also easily deforms during the finishing process due to its flexibili ty. These factors increase the difficulty of achieving a smooth internal finish on flexible capillary tubes. This research develops a new finishing method featuring the use of a multiple (double) pole tip system with a special magnetic tool This method pr oduces smooth inner surface s of capillary tubes and finishes multiple areas simultaneously to improve finishing efficiency. This study explains the processing principle s of newly developed finishing method and describes the surface finishing characteristic s. The finishing mechanism i s also thoroughly investigated and described in both straight and flexible capillary tubes
14 CHAPTER 1 INTRODUCTION 1.1 Background 1.1.1 Introduction of Flexible Capillary Tubes and Necessity of Internal Clean ing Capillary tub es are used for refrigerant conduction in industrial refrigerators [ 1 ] in needles for biopsy procedures and in precision analytical instruments Flexible capillar y tubes are widely used for medical applications such as coronary stents and catheter shafts The components for medical applications are remarkably precise and sensitive, but can be fatal to patients when manufactured with any defects or misalignment during installation [ 2 3 ]. M edical stent s are implanted into arteries in order to expand blood vessels and guarantee blood flow, and catheter shafts guide other medical components into the human body Currently, these devices are made of 304 and 304L stainless steels which are widely available, cost efficient, have good corrosion resistance and ha ve high machinability and weldability which allows them to be drawn and formed into the required long fine tubes (304L is good for welding because it has 0.03 % ( max ) carbon content which eliminates carbide precipitation due to welding , whereas 304 has a higher of carbon level (0.08 % max)). T he minimum tube size for medical usage is 0.005 ( 0.127 mm ) outer diameter (OD) and 0 .00325 ( 0.082 mm ) inner diameter (ID). T o fit into the human body, the components must have inherent flexibility, which is a ccomplished by numerous laser etched slits into the tube wall by laser micromachining processes [5, 6]. A hypotube a critical component of the balloon catheter is a long metal tube with micro engineered features along its length. This tube enables a card iologist to navigate the vascular anatomy to the site of the clogged artery and ensure accurate placement of
15 the balloon and stent. To achieve optimal performance, the factors pushability, torqueability, kink, and trackability of the tube need to be consid ered during design Pushability is the ability of the shaft to transmit energy from one end of the catheter to the other, and this can be improved by increasing the wall thickness of tube, reducing its overall length, or increasing the stiffness of the mat erial. Torqueability is the ability of the shaft to transmit a rotational displacement along the length of the shaft, and it can be improved by increasing the wall thickness or shear stiffness of the material. A measure of a shaft s ability to maintain its cross sectional profile during deformation is known as kink A bending or compressive force on shaft causes kink failure. Trackability is the ability of a shaft to travel or track through tortuous anatomical structures. Trackability is influenced by the f lexibility of the shaft and can be improved by reducing the shaft s outer diameter or decreasing the material s elastic modulus. These factors are controlled by patterns of multiple sl i ts made in the tube walls by the laser micro machinin g process . Lase r micromachining processes are generally used for cutting, drilling, marking, texturing, welding, and developing for medical (stents, shafts) optical (micro lens) automotive (fuel injector nozzle, high pressure fuel filter) aerospace (nano satellite) an d microelectronics industry (silicon micromachining) applications Laser micromachining processes cut material by melting the workpiece with thermal energy and are capable of making small holes and slits in capillary tubes [8, 9] T he se multiple slits in t he tube wall increase the flexibility of the tube, as is desired in components such as catheter shaft During the laser machining process some molten material splashes inside the tube, solidifie s and adhere s to the inner surface or cut edges and re soli dified material
16 form s hard heat affected burrs . The burrs cause serious problems in manufacturing and t he quality of precision parts can be evaluated by the surface and edge quality ; therefore they must be removed completely in a subsequent process to allow the proper operation of components T he existence of burr s in parts deteriorates the accuracy and the performance of the products. However, it is difficult to use existing conventional techniques to remove the hard heat affected burrs and re solid ified material that project inside the capillary tubes. 1. 1 2 Current Techniques for Internal Machining of Tubes and Magnetic Abrasive Finishing For internal deburring of flexible capillary tubes, c onventional deburring tools cannot be easily introduce d or adequately controlled inside flexible capillary tubes that have inner diameters around 1 mm and incorporate micro machined multiple slits. Chemical and micro abrasive blasting processes are commonly used for deburring flexible tubes [11 13]. However, it i s difficult to drag deburring tools, including chemical reagents and abrasive media, inside long flexible capillary tubes. I t is difficult to remove the burrs completely even using these nontraditional techniques. Moreover, t he removal of solidified materi al adhered to inner tube surface s and cut edges necessitates machining harder material, which i s a result of the laser induced heat affected zones. In the case of internal surface finishing, the variations in the as received peak to valley surface roughnes s can be several micrometers In contrast, the burr height can be up to 100 m, and therefore the deburring process requires greater material removal than the internal surface finishing  In surface finishing, traditional machining process leaves impur ities such as rust spots, scale and non metallic inclusions on the grooves of the machined surface which
17 are unacceptable for proper performance . To reduce these defects, nontraditional finishing methods such as electropolishing [16 19], known as el ectrochemical polishing or electrolytic polishing has been considered and developed. Figure 1 1 Schematic of electropolishing process Electropolishing is a chemical surface finishing technique applied to clean metallic surface s to reduce micro roughn ess The process depends on electrolysis which uses a flow of current and a solution comprised mostly of sulphuric and orthophosphoric acid s as an electrolyte This process removes metal from a submerged workpiece by passing an electric current through th e electrolyte It can produce a smooth, bright, and reflective surface that exhibits superior corrosion resistance when the workpiece ( anode ) and tool electrode ( cathode ) are electrically charged (Figure 1 1) Electropolishing is suitable for the polishing of both complex shapes and hardened materials, which are difficult to machine mechanically because the electrode and the workpiece are not in contact with each other [15, 20 ] Electropolishing produces mirror like surfaces on metals as well as being used for deburring by electrolyte which removes material from peaks and raised area faster than lower areas. However, disposal of the chemical reagents after the process can contribute to the destruction of
18 the environment In the case of capillary tube s it is difficult to introduce the electrolyte into the tubes T herefore an alternative process is desired. Recent developments in nonconventional machining process technology have enabled automated micro deburring and surface finishing for precision parts. They include ultrasonic vibratory finishing [2 1 ], ultrasonic cavitation deburring [2 2 ], thermal energy deburring [2 3 2 4 ], abrasive flow machining [2 5 2 7 ], laser deburring [2 8 ], magnetic barreling [ 29 ], and magnetic abrasive finishing [3 0 3 3 ]. However, for fin ishing the inner surface s of capillary tubes, only abrasive flow machining and magnetic abrasive finishing are potential ly viable processes Abrasive flow machining p erforms deburring of the small holes of diesel injection nozzles made by drilling electro discha r ge machining, or laser machining. The pressurized media (a mixture of a polymer carrier, lubricant, and abrasive ) pass through the holes and finish the inner surface s, as well as remove the burrs. However, a s long as the abrasive behavior is control led by media pressure at the end of the hole the pressure drop along the length of the hole will impede the expansion of the process application [2 5 2 7 ] Although this process is proper for surface finishing of capillary tubes, it may cause difficulties f or internal deburring of flexible capillary tubes due to the significant pressure leakage from the slits. In internal finishing by magnetic abrasive finishing the abrasive mixed with ferrous particles is introduced into a tube and pushed against the inner surface of the tube by magnetic force in the presence of a magnetic field The smooth relative motion of the magnetic abrasive against the inner surface results in material removal and successful surface finishing when the tube rotates Manipulating the m agnetic field
19 along the tube axis direction drags the magnetic abrasive to target finishing area and achieves the internal finishing of long tube s [3 3 ]. A study of the internal finishing of straight capillary tubes has demonstrated successfu l surface finis h ing of tube s with inner diameter s down to 400 m  ; h owever, the machining rate of this process was low The internal deburring and finishing efficiency of flexible capillary tubes have yet to be investigated 1.2 Development of Internal Cleaning and F inishing Process for Flexible Capillary Tubes using Magnetic Abrasive Finishing The objectives of this study are to develop a new internal surface and edge finishing method using the magnetic abrasive finishing process and to apply the process to flexible capillary tube s In order to achieve these goa l s three sp ecific aims are proposed : 1. Determine fundamental finishing characteristics and mechanism of flexible capillary tubes in magnetic abrasive finishing 2. Develop a new finishing method using a heat treate d metastable austenitic stainless steel tool in combination with multiple pole tips. 3. Design and develop a high speed finishing machine and determine finishing characteristics and mechanism s of both straight and flexible capillary tubes using the multiple pole tip system T o achieve successful finishing of capillaries by magnetic abrasive finishing, controlling the magnetic abrasive and lubricant at the finishing area inside the capillary tube is critical This can be accomplished by controlling the magneti c field generated by the permanent magnets through the modification of the magnetic pole tip geometry W hen the tube diameter decreases, the width of pole tip along the tube axial direction is shortened. This causes a shortened default finishing length and results in increased processing time. Moreover, in deburring of flexible capillaries, the burrs on the edge
20 impede the movement in the tube axial direction of the particles introduced into the tube This limits the pole feed length and reduces finishing e fficiency. Moreover, to remove the hard material on the machined edge ( the area s hard to reach by conventional method s) a large sized diamond abrasive is required to enhance the material removal rate , and magnetic tools are applied inside the tube to strengthen the magnetic force using the volume of ferrous tool and to keep the flexible tube straight during the process at the finishing area [3 3 ]. B ased on these methods, a new finishing process to have multiple processing areas for a single tube and to process multiple tubes simultaneously has been considered t o improve the processing rate per tube [3 5 38 ]. It is called a multiple pole tip system To realize this concept, a partially heat treated metastable austenitic stainless steel tool has been fabri cated. The roles of ferrous tools, the effects of their geometric and magnetic properties, and the dynamic behaviors on finishing characteristic for capillary tube finishing are studied. Another key factor for i mproving the processing rate is to increas e t he tube rotational speed P reston introduced an empirical fo r mula for determining how quickly material is removed from a surface . According to Preston s equation, the material removal rate is a function of the finishing pressure or force per unit area relative speed between the tube and abrasive, and the Preston coefficient, whi ch is determined by the finishing system ( e.g. by the abrasive type and size, and friction coefficient between the abrasive and tube ) The magnetic force is the main parameter determining the finishing pressure. The relative speed between the tube and abrasive is controlled by the combination of tube rotational speed and pole reciprocation speed in the direction of the tube axis. The h igher the relative speed, the longer the th eoretical sliding distance of the
21 abrasive against the tube surface. A p revious study has shown the feasibility of high speed finishing machine that can rotate the tube up to 30000 min 1 . A new high speed machine has been designed and manufactured wit h a spindle speed up to 30000 min 1 The developed machine has shown its finishing capability successfully, and t he effects of tube rotational speed on abrasive motion and finishing characteristics using capillary tubes are studied. For further improvement of finishing efficiency, the multiple pole tip system for the high speed machine is proposed and implemented Based on high speed finishing experimental results, the finishing characteristics and mechanism using a high speed multiple pole tip system for s traight and flexible capillary tubes will be studi ed and clarified
22 CHAPTER 2 INTERNAL FINISHING P ROCESS USING MAGNETI C ABRASIVE FINISHING 2.1 Magnetic Abrasive Finishing Magnetic abrasive finishing is defined as a finishing process that removes material in the presence of a magnetic field. This process was invented during the 1930s but not further developed until after the 1960s [3 1 40], and it has been developed as a new finishing technology in the last decade and still a useful and viable finishing me thod. In the last decade, the magnetic field assisted finishing process has garnered much attention; it is widely used in the field of precise and sensitive instrument manufacturing such as for medical, optics, electrical and engine components [41 45]. The magnetic field assisted finishing process is based on the use of slurry consisting of ferrous particles mixed with fine abrasive particles ( d iamond a luminum o xide (Al 2 O 3 ), s ilicon c arbide (SiC), c ubic b oron n itride (CBN) etc.). These non metallic abrasi ves have hard sharp edges with irregular shape s This mixture combine s along the line s of the magnetic field ( generated by poles attached to magnets ) and forms a flexible magnetic abrasive brush. T he flexibility of the brush can be altered by adjusting the magnetic field generated by a permanent or alternating magnet system, the brush performs like a multi point cutting tool for finishing process because it has the ability to modify itself according to surface contours [46 47 ]. In a magnetic field strong en ough to overcome the inherent friction between the abrasive and a target surface, t he motion s of the abrasive brush finish the surface In this process, ferromagnetic particles sintered with abovementioned non metallic abrasives (diamond, Al 2 O 3 SiC,or CBN ) and such particles are called magnetic abrasive
23 In a magnetic field, magnetic flux flows unimpeded through nonferrous workpiece material, and ferrous material a component of the magnetic tool is suspended by magnetic force. It is possible to influence t he magnetic tool motion by controlling the magnetic field, thus enabling the finishing operation to be performed not only on easily accessible surfaces but also on areas that are hard to reach by conventional mechanical techniques. Figure 2 1 Schematic of the processing principle for plan ar magnetic abrasive finishing Figure 2 1 shows a schematic of the processing principle for plan ar magnetic abrasive finishing process where the finishing action is generated by application of a magnetic field across the gap between the surface of the workpiece and a rotating magnetic pole. The magnetic field generated by the magnetic pole forms the (self adaptive) magnetic abrasive brush, and the normal force acting on the workpiece surface, in combination with pole rota tion, causes material removal ( in the form of chips ) and gradually improves the surface roughness with pole feed
24 Figure 2 2 Schematic of the processing principle for cylindrical magnetic abrasive finishing Figure 2 2 shows a schematic of the processin g principle for a cylindrical magnetic abrasive finishing process. The magnetic abrasives are joined magnetically between magnetic poles (N and S) along the lines of magnetic force and form flexible magnetic brush. When the rotating cylindrical workpiece i s introduced into the magnetic field between the poles, surface and edge finishing are performed by the magnetic abrasive brush [3 1 ]. The magnetic abrasive finishing process has certain advantages which make it an efficient process: 1. The abrasive brush is f lexible to conform to workpiece surface and hence complex surfaces can be finished. 2. The finishing pressure can be controlled by varying the magnetic field (current in the case of electromagnet and air gap in the case of permanent magnet). 3. The finishing too l is independent. 4. In electromagnet system, the disposal of used abrasive is automatic and new abrasive is fed to the finishing area by turning the current in the coils on and off. 5. No scattering of abrasives due to the magnetic field. 6. Less consumption of ab rasive
25 2.2 Processing Principle and Parameters of Internal Finishing of Straight Tube Internal finishing by MAF has two system configurations One is the rotating tube stationary pole system [ 33 ], and the other is the rotating pole stationary tube system . These systems are chosen to suit the workpiece geometry. The former system is suitable for short workpiece s that are rotatable at high speeds. The latter system was developed for non rotatable workpieces, which have long, large or non rotation ally sy mmetric geometry, such as elbows, bent tubes, and slender tubes [ 48 ] Since the workpiece s provided for this study were relatively short ( approx. 100 mm long) and rotatable at high speeds, the rotating tube station ary pole system was selected Figure 2 3 Schematic of processing principle of internal finishing of straight tube using rotating tube stationary pole system Figure 2 3 shows a schematic of the internal finishing process using the rotating tube stationary pole system for tubes. The desired ma gnetic field in the finishing area is generated by permanent magnets attached to a steel yoke. In the case of capillary tubes, to concentrate the magnetic field in specific area, a tapered steel pole tip is used. In the presence of a magnetic field, magnet ic abrasive introduce d inside the workpiece
26 is magnetized, and it is pushed against the workpiece inner surface by magnetic force. In a non uniform magnetic field, the magnetic force F acting on magnetic abrasive can be described using following equation [ 36 ] (2 1) where V is the volume of the magnetic abrasive, is the susceptibility, and H and grad H are the intensity and gradient of the magnetic field, respectively. If the tangential component of the magnetic force acting on the magnetic abrasive is larger than the friction force between the magnetic abrasive and the inner surface of the workpiece, the magnet ic abrasive shows smooth relative motion against the inner surface when the workpiece is rotated at high speed [3 4 ] Ma nipulating the poles along the workpiece axis causes the magnetic abrasive to move in the axial direction following the pole motion, eff e ctively finishing the inner surface and removing burrs in the case of flexible capillary tubes. The clearance between the pole tip and tube should be set at approximately 0.1 mm; this can be accomplished by using p olytetrafluoroethylene (PTFE) tape on the pole tip surface. In the case of capillary tube case, magnetic force attracts the tube towards pole tips and they are always contacted each other. This tape also prevents damage to the outer surface of workpiece from the pole tip during rotation In addi tion in the case of flexible capillary tubes which have spiral slits, rotating t he tube in the direction in which the motion itself acts to close spiral slit is chosen This helps not to lose the abrasive and lubricant from the slits. As the tube is rotat ed, the mixture of ferrous particles exhibits relative motion against the tube inner surface and thus material is remov ed from the tube surface. The default finished length (the finished
27 length without pole tip motion) is determined by the pole tip width in the tube axial directio n ; motion of the pole along the tube axis extends the finished area. A Iron particle B White alumina magnetic abrasive Figure 2 4 Microscopy and scanning electron microscopy of A) iron particle (dia. 150~300 m) and B) white alumina (WA) magnetic abrasive (mean dia. 80 m, a luminum oxide (Al 2 O 3 ): 10 m) This process controls magnetic abrasive chains where the abrasive mixtures conform to the contour of the surface to be finished. The mixture of iron particle s (Figure 2 4 (A ) ) and aluminum oxide magnetic abrasive (Figure 2 4( B )) is called mixed typ e magnetic abrasive and is introduced into the tube. The magnetic abrasive is pushed by iron particle s and held firmly against the inner surface of workpiece while short reciprocating motion s are applied in the workpiece axial direction. The mixed type 1 mm 1 mm 100 m 25 m
28 mag netic abrasive contacts and acts upon the surface When removing the surface defects such as scratches, lay lines and other disparities, these defects on the surface are corrected to a limited depth of few microns [4 9 ]. The material removal rate depends o n the relative motion of magnetic abrasive, which depend s on the magnetic force, workpiece rotational speed, abrasive size and type, working clearance workpiece material, feed rate, and lubricant. To achieve smooth relative motion of the mixed type magnet ic abrasive, it requires that the mixed type magnetic abrasive must be kept in a stable configuration during rotation in the finishing area against the inner surface of the tube. This relative motion determines the material removal and results in successfu l surface finishing. To keep the mixed type magnetic abrasive in the finishing area, a strong magnetic force is required, which is determined by the magnetic field generated by magnet. Based on the magnetic force equation (2 1), magnetic force acting on th e abrasive depends on the volume of ferrous particles. Due to the limit ed i nner space in capillary tubes, the size of iron particle s has a maximum constraint and the volume of mixed type magnetic abrasive must be close to 43 vol% of the tube for efficient finishing . The tube rotational speed also affects smooth relative motion of the magnetic abrasive against the inner surface. Preston s equation predicts that material removal increase s with increased rotational speed. However, at high rotational speed s, lubricant can be easily spun out from the finishing area due to centrifugal and frictio nal force s The lubricant encourages the mixed type magnetic abrasive to congregate uniformly on the inner surface of the tube at the beginning of the finishing proce ss. Moreover, lubricant helps smooth relative motion of particles against the inner surface and
29 removes the heat, chips, grain fragments and dislodged grains from the surface during the finishing process. Therefore a lack of lubricant deteriorates the re lative motion and material removal. The overall finishing efficiency of the process is dependent on the default finishing length, which is determined by the width of the pole tip along the tube axis direction. T he m agnetic flux density is stronger at the e dges of pole tip because of edge effect s This condition causes powder to be distributed non uniformly on the finishing area because it gives ferrous mixtures a higher tendency to collect towards one edge ( usually the edge closest to where the mixture is i ntroduced ) instead of uniform distribution in the finishing area. As a result, the abrasive relative motion is obstructed and this results in unsuccessful surface finishing. This causes the pole tip width to be shortened in capillary tube finishing and inc reases the processing time. C ompared to a surface finishing process an internal deburring process requires a higher material removal rate This requirement is met by increasing the intensity of the magnetic force acting on the magnetic abrasive or employi ng larger abrasive sizes In this study, in order to increase material removal with a strengthened magnetic force, (non ferrous) diamond abrasive is used and a solid ferrous rod is substituted for a portion of the ferrous particles.
30 CHAPTER 3 PARAMETER S AFFECTING ON FINSI HNG CHARACTERISTICS IN FLEXIBLE CAPILLARY TUBE FINIS HING USING A SINGLE POLE TIP SYSTEM 3.1 Processing Principle of Flexible Tube Finishing A schematic of internal machining process of flexible capillary tubes using a rotating tube stat ionary pole system is shown in F igure 3 1. The desired magnetic field in the finishing area is generated by permanent magnets attached to the steel yoke ( F igure 2 2). Since laser machined burrs have various shapes and heat affected zones, sharp and hard cu tting edges combined with strong magnetic force are required to remove the burrs. Increased magnetic force using large sized ferrous tool contributes to the remov al of material adhered to the tube inner surface and burrs along the slits ( shown in F igure 3 2( A ) and ( B ) burr heights: up to 9 0 m ) The internal volume of the capillary tube is limited, so there is an upper constraint to the size of the introduced ferrous materials. In an attempt to encourage the cutting action large sized diamond abrasive ( 4 8, 20 40, 50 70 m diameter ) was introduced A mixture of iron particles, magnetic abrasive, and diamond abrasive is attracted to the finishing area by the magnetic field generate d by permanent magnets. The sizes of these particle s must be greater than the slit widths to prevent leakage Manipulating the poles along the tube axis causes the mixture to follow the pole motion and move in the axial direction, finishing the inner surface and removing the burrs The methods without the modification of finishing unit to adjust the magnetic field are to change magnetic properties or geometry of the magnetic abrasive [ 44, 4 6 ,50,51 ], or to change magnetic properties, geometry, or concentration of magnetic tools (such as ferrous particles, a ferrous rod, or permanent m agnet) inserted with magnetic abrasive or conventional abrasive slurry [ 52 5 7 ].
31 Figure 3 1 Schematic of processing principle of internal finishing of flexible tube Figure 3 2 Exterior and interior microscopy of laser machined flexible capillary tubes Tube s A and B In this study, two different flexible capillaries are prepared. In F igure 3 2, exterior and interior microscopy of tube A and B which have different diameters and slit patterns
32 machined with laser on the body are shown. Tube A has 0.58 mm O D, 0.42 mm ID and spiral slit pattern on it. The slit has 12 m distances and this can be wider or narrower according to the tube rotational direction and it affects total tube length. Tube B has a 1.36 mm OD and 1.02 mm ID, it has more numbers of slit whi ch make the tube more flexible than tube A. In Figure 3 2, the edges have burrs; moreover re solidified material caused by thermal energy during laser machining exists between the slits. Figure 3 3 Photographs of tubes chucked on the finishing spindle w ith different slit patterns, Tube s A and B (Photograph courtesy of Junmo Kang) Figure 3 3 is photographs of tube A and B chucked on the finishing spindle respectively. One end of tubes is chucked on finishing unit while other end is free. Tube A with a few slits can be treated as a rigid tube as shown in Figure 3 3( A ), and more slits encourage the tube flexibility as shown in Figure 3 3( B ), it becomes no longer to treat it as a straight rigid tube. Even the other end is supported flexible jig, tube fluctuat e s during it rotates at high speed when the tube length gets longer. Moreover, abrasive and lubricant can leak easily from the slits. W ithout critical control of magnetic abrasive and lubricant at the finishing area, it is unfeasible to produce the smoothl y finished surface and the deburred edges.
33 The internal surface and edge finishing experiments with tube A ( F igure 3 2( A )) have been conducted focusing on mostly diamond abrasive size and processing time. Due to small (less than 10 m) cutting edges embed ded in magnetic abrasive, it has difficulties to remove material of the burrs inside the tube which is not suitable. To remove large sized burr (50 90 m height), diamond abrasives are introduced during the finishing process. In case of tube B ( F igure 3 2( B )), even the tube is allowed to introduce more ferrous particles inside, the laser machined slits are closer each other and they are formed along to cutting direction (same as tube rotational direction), which makes the difficulties to be removed. Theref ore two kinds of magnetic tools (steel rod and stainless steel rod) have been considered to increase the magnetic force acting on the magnetic abrasive to go over the burrs. Using single pole tip system, the factors effects on surface and edge finishing o f flexible capillary tubes, and the roles and behaviors of magnetic tools inside the tubes will be discussed. 3.2 Effects of Finishing Time, Tube Rotation, Slits, and Abrasive Size on Edge and Surface Finishing 3.2.1 Experimental Setup and Conditions A n e xternal view of the experimental setup for flexible capillary tube finishing is shown in F igure 3 4 The stainless steel capillary tube with spiral slits (Tube A) is held by the chuck While the tube is rotated, centrifugal force acts inside the tube Depe nding on the direction of the tube rotation, the tube either elongates or shortens by changing the slit widths because of its spiral slits The rotational direction is determined with the direction make the tube shortened to keep the abrasive and lubricant inside the tube. T o diminish the run out of the capillary tube beyond the finishing area the other end of tube is held by a flexible jig which also can react to the changes in the tube length.
34 Figure 3 4 External view of finishing equipment for flex ible capillary tube finishing (Photograph courtesy of Junmo Kang) Three n eodymium i ron b oron (Nd Fe B) p ermanent magnets (10 12 18 mm, r e sidual flux density 1284 mT, c oercive f orce 1440 kA/m) generate the magnetic field needed for attracting the magnetic a brasive to the finishing area. The magnets can be oscillated in the axial direction by a crank mechanism connected to the motor and also fed in the axial direction of the tube with or without oscillation. T he pole tip geometry shown in Table 3 1 was deter mined based on the previous study of capillary tube finishing , and the pole tip surface was covered by 0.11 mm thick polytetrafluoroethylene (PTFE) tape. The tube was in contact with the pole tip surface during finishing This minimize d the clearance between the pole tip and finishing area to maintain a strong magnetic field at the finishing area and contributed to diminishing the run out of the tube during finishing. The tape also protect s both pole tip and the tube from rubbing against each other.
35 T he previous research  reported that the key to achieve finishing of capillary tubes was the combination of the methods that a) C ontrol the magnetic field at the finishing area b) D etermine the appropriate amount of mixed type magnetic abrasive and c) S upport the capillary at three points T he finishing setup and conditions were refined t o satisfy the abovementioned requirements Table 3 1 shows the experimental conditions. 304 stainless steel tubes ( 0.42 mm ID 0.58 mm OD and 55 mm length ) with spir al slits were provided as workpieces for this study. The undeformed sl i t width was measured at about 12 m The tube was chucked 12.5 mm from one end Since l a ser machined burrs have various irregular shapes and heat affected zone s sharp and hard cutting edges combined with strong magnetic force are required to remove the burrs. C onventionally used for 304 stainless steel tube finishing m ixed type magnetic abrasive a mixture of relatively large sized iron powders and white alumina magnetic abrasive (mean diameter 80 m and comprising iron particles and Al 2 O 3 abrasive grains under 10 m in diameter) could not remove the heat affected hard burrs. D iamond abrasive in a paste form was, thus, added to the mixed type magnetic abrasive with lubricant. The WA mag netic abrasive has irregular microasperities on the surface produced during the manufacturing process. These microasperities allow the WA magnetic abrasive to hold the nonferrous diamond abrasives through the lubricant at the finishing area. One of the imp ortant properties of the lubricant must be the viscosity, which must be high enough to prevent leaking through the slits and to encourage the lubrication between the abrasive cutting edges and target surface b ut low enough to be introduced into the capilla ry tube.
36 Table 3 1. Experimental conditions for internal deburring of flexible capillary tube with laser machined slits : Tube A Workpiece Tube A 304 stainless steel tube ( 0.58 0. 42 55 mm) Workpiece revolution 2500 min 1 Ferrous particle Iron particles ( 50/+100 mesh, 150 300 m dia.): 80 wt% + Aluminum oxide (WA) particles (80 m mean dia.) in magnetic abrasive (<10 m): 20 wt% Amount of ferrous particle 0.5 mg Diamond abrasive 4 8, 20 40, 50 70 m diameter Lubricant Soluble type barrel finishing c o mpound: 4.53 L Permanent magnet Nd Fe B permanent magnet : 10 12 18 mm Pole reciprocating motion Amplitude: 4 mm, Speed: 0.6 9 mm/s Workpiece Pole clearance 0.11 mm Pole tip 1080 carbon steel The necessary amount of the magnetic abrasive can be determin ed according to the ratio of the volume of the magnetic abrasive to t he overall volume inside the tube (in a length corresponding to the width of the pole tip) There exists a threshold of the supplied abrasive amount between 43 and 55 vol% ; a s upplied amount less than the threshold is required for efficient finishing . The mixture of 0.1 mg WA magnetic abrasive and 0.4 mg iron particles which results in 47 vol% taken up by the iron powder and WA magnetic abrasive was supplied for the experim ents Under the conditions with high speed tube rotation the mixture of the magnetic abrasive and iron particles shows difficulties to follow the pole s feed in the tube axis
37 direction because the friction force becomes much larger than the magnetic force The burrs also obstruct the mixture to follow the poles feed. Therefore the feed speed was set at the lowest of the experimental setup, 0.6 9 mm/s. The crossing angle of the cutting marks generated by the diamond abrasive is calculated as 2 tan 1 ( v f / v r ), where v f is the pole feed velocity and v r is the tube rotational speed. Under the experimental conditions, was calculated to be 1.4 T he effects of the diamond ab rasive size on the processing characteristic s were examined. Three sizes of diamond abrasive were prepared : 4 8, 20 40, and 50 70 m which are smaller, slightly larger, and much larger than the slit width respectively The finishing unit was reciprocated over 4 mm at 0.6 9 mm/s, finishing an 8 mm length located 3 mm from the end opposite the chuck T he stylus of the surface roughness profilometer which is the device most commonly used for surface roughness measurement, could not be inserted inside the 0. 42 mm ID capillary because of the small tube diameter. This created difficulty in measuring and tracking the changes in the surface roughness and burr heights with finishing time. The tube was cleaned using ethanol (200 proof) in an ultrasonic cleaner ever y 6 min, and the changes in the material removal with finishing time were tracked only by measuring the weight reduction by the process with a micro balance (10 g resolution ). Following replenishment of the mixture of the ferrous pow d ers and abrasive ever y 6 min t he finishing experiments were continued. After the finishing for a certain period, the tube was sectioned along the tube axis, and the surface was evaluated using an optical profilomet er and microscop e
38 3.2.2 Effects of Diamond Abrasive Size on D eburring Flexible Capillary T ube The time sufficient for finishing was initially determined by examining the changes in the finishing characteristics with finishing time using the 50 70 m diameter abrasive. The diamond abrasive is larger than the undeform ed slit width, measured to be 12 m Figure 3 5 shows the micrographs of the unfinished surface and the surface finished for 30 min and 60 min. Although areas damaged by the laser machining process are observed between the slits, the initial surface burrs adjacent to the slits are the target of this study. The initial height s of the burrs generated by laser machining were measured by an optical profiler to be in the range between 8 m and 50 m During the process, the tube was deformed such that the slit s were narrowed due to the tube rotation. When the diamond abrasive encounters the slit during the process, the abrasive momentarily loses contact with the surface but contacts the surface again on the other side According to the cutting marks, the reeng agement of the abrasive must be smooth. The diamond abrasive must gradually remove material from the peaks of the burrs and the microasperities of the tube surface. The conditions for 30 min were found to be insufficient for effective deburring. By increas ing the finishing time, however, the edge and surface finishing were both smoothly performed. Figure 3 6 shows the changes in the material removal with finishing time. The variation of the material removal at the beginning was due to the variation in the i nitial burr conditions. T his indicated that about 0.3 mg of material removal might be necessary for both edge and surface finishing of the tubes for 8 mm finishing length Accordingly, it was shown that MAF is applicable for inner surface finishing and rem ov al of laser machined spiral burrs projected inside flexible capillary tubes. Next, the effects
39 of the diamond abrasive size on the deburring characteristics were examined with three sizes of diamond abrasives : 4 8, 20 40, and 50 70 m diameter Figure 3 5 Micrographs of unfinished surface and surfaces finished for 30 and 60 min with 50 70 m diamond abrasive Figure 3 7 shows micrographs of the surface s finished with 20 40 and 4 8 m diamond abrasive for 60 min. In the case with 20 40 m diamond abrasi ve, shown in Figure 3 7( A ), the burrs and the initial surface unevenness remain due to a lack of material removal. This is because of the smaller cutting edges of the diamond
40 compared to the 50 70 m diamond abrasive. In the case of the 4 8 m diamond abra sive, shown in Figure 3 7( B ), no burrs remain, but deep irregular undulations are observed on the finished surface. During the process, the magnetic force pushes the iron particles, which push the WA magnetic abrasive which, in turn, push the diamond abr asive against the inner surface of the tube. Some 4 8 m diamond abrasive must agglomerate between the WA magnetic abrasives and tube surface. The mass of the diamond abrasive pushed by the WA magnetic abrasive and ferrous particles must participate in the finishing performance, resulting in the deep scratches. This also led to the remov al of the burrs and surface finishing with irregular undulation s Figure 3 6 Changes in material removal with finishing time
41 Figure 3 7 Micrographs of surface finis hed with 20 40 m diamond abrasive and 4 8 m diamond abrasive Figures 3 8 and 3 9 show the three dimensional shape s ( measured by an optical profiler ) of the unfinished inner surface and the inner surface s finished with three sizes of diamond abrasives fo r 60 min. In F ig ure 3 8 the surface data in the area of a slit is passed through a low pass filter so that the burr shapes can be examine d The irregularity of the surface brought about measurement difficulties, resulting in the discontinuity of the surfa ce seen in F ig ure 3 8 However, the surface is continuous except for the slit. The abovementioned burr height s were obtained as peak to valley value s of l ines drawn perpendicular to the slit from several observations. To examine the surface roughness Fig u re 3 9 shows the un filter ed area between slits. As shown in Fig ure 3 8 ( A ) and 3 9 ( A ), the case with 50 70 m diamond abrasive demonstrated the
42 burr removal and surface finishing. T he surface was modified from 0.67 to 0.12 m Ra In the case of the 20 40 m diamond abrasive, the initial burrs remain ed after 60 min, as shown in Fig ure 3 8 ( B ). This condition must have merely removed material from the peaks of the surface irregularities and left the 5 7 m high burrs. Although the three dimensional surface shap e observation barely show s the initial micro unevenness remaining the small material removal resulted in slowed surface roughness improvement from 0.65 m 0.21 m Ra ( Figure 3 9( B ) ) However, t he 4 8 m diamond abrasive removed burrs successfully, as sho wn in Figure 3 8( C ). The surface finished by the 4 8 m diamond abrasive consists of the accumulation of shallow cutting marks on the deep undulations with relatively longer wavelength which must be caused by the agglomerated diamond abrasive. This genera ted the surface roughness value to 0.35 m Ra (Figure 3 9( C )). The cases with 4 8 and 20 40 m diamond abrasives show an other trend as well After finishing a mixture of lubricant, diamond abrasive, and chips was observed outside the tube, and cutting mar ks were observed on the outer surface of the tube as shown in Figure 3 10. The heat generated during the finishing process must decrease the v iscosity of the lubricant, and the 4 8 m diamond abrasives which are smaller than the slits, must have leaked t hrough the slits. The diamond abrasive bec ame sandwiched between the rotating tube and pole tip, and achieved external surface finishing of the tube. The unagglomerated 4 8 m diamond abrasive must pass easily through the slits, and a large number of abras ive must participate in finishing of the outer surface. On the other hand, the 20 40 m diamond abrasive could have been larger than the slit widths. In practic e a distribution including smaller sized diamond abrasive results from the
43 manufacturing proces s. Alternatively the diamond abrasive could be crushed during the process, and some abrasive would thus be reduced in size The diamond abrasive that was able to migrate from the internal finishing area must have caused the scratches on the outer surface and slightly removed the material from outer surface. Figure 3 8 Three dimensional burr shapes measured by optical profiler ; A) edge finished by 50 70 m diamond abrasive B) edge finished by 20 40 m diamond abrasive and C) edge finished by 4 8 m diamond abr asive
44 Figure 3 9 Three dimensional surface shapes measured by optical profiler ; A) surface finished by 50 70 m diamond abrasive B) surface finished by 20 40 m diamond abrasive and C) surface finished by 4 8 m diamond abrasive
45 Figure 3 10 Micrographs of outer surface of tube; A) as received surface, B) surface finished with 4 8 m diamond abrasive and C) surface finished with 20 40 m diamond abrasive Figure 3 11 Changes in material removal with finishing time according to diamond abrasive size Figure 3 11 shows changes in the material removal with finishing time. The material removal in the 20 40 m diamond abrasive case is lowest of the three conditions. It is noted that the material removal in the 4 8 and 20 40 m diamond
46 abrasive cases are the results o f finishing of both the inner and outer surface s In the case with 20 40 m diamond abrasive, the lack of large cutting edges of the abrasive resulted in the smallest material removal. R egardless of the fact of being the smallest abrasive of the three the effects of t he agglomeration of the 4 8 m diamond abrasive in the internal finishing and the leakage of the diamond abrasive from the slit for the outer finishing led to the greatest material removal. 3.3 Roles and Behaviors of Magnetic Tools In an att empt to encourage the material removal rate for flexible capillary tubes diamond abrasive was introduced to have large cutting edges This method was successful for tube A with less number of slits and small burrs (up to 50 m). However, it was unsuccessf ul in finishing trial for tube B with more number of slits and large burrs (up to 90 m), because the abrasive size is smaller than the burr height, that is, abrasives cannot go over the burr when the magnet moves along the tube axis (Figure 3 12). Therefo re, stronger magnetic force to overcome these difficulties is required. B ased on the magnetic force equation (Eqn. 2 1), change the volume of ferrous particle is considered. Increasing the diameter of the iron particles is not the only method for increas i ng the volume of included ferrous tools Increasing the length of the ferrous tool along the tube axis direction, that is replacing the some ferrous particle with a rod is another method to increase th e volume and thus increas e the magnetic force acting o n magnetic abrasive (Figure 3 13) The geometric as well as magnetic, propert ies of the ferrous tools play important roles in realizing the MAF process in the limited space especially capillary tube s
47 Figure 3 12 Schematic of iron p article, magnetic a brasive, and diamond abrasive behavior inside the flexible capillary tube with magnet feed Figure 3 13 Schematics of processing principle with different magnetic tools (switching the iron particle to rod) Previously, carbon steel (JIS designation : S48 C ( 0.48 wt% C ) ) and austenitic stainless steel rods or pins replaced iron particles for the surface finishing of 18 mm ID tube [5 8 ]. The carbon steel is ferromagnetic and exhibits high magnetic susceptibility. Austenitic 304 stainless steel is paramagnetic but becomes f erromagnetic when plastic deformation causes the face centered cubic (austenitic, nonferrous) structure to be transformed into a body centered cubic ( martensit ic, ferrous) structure. Moreover, its
48 magnetic anisotropy is attributed to the shap e of martensite formed in the austenitic matrix during the plastic deformation [5 9 6 3 ]. When the same mass of stainless steel rods, carbon steel rods, or iron particles is subjected to the same magnetic field, the magnetic force generated with the stainles s steel rods is about three to four times lower than the force with carbon steel rods and about half of that with iron particles [ 54 ]. Introducing the carbon steel rods into the finishing area generated higher magnetic force than the iron particles. In the magnetic field, t he steel rods ( Figure 3 14( A ), 0.5 5 mm) are magnetized in line with the magnetic flux (i.e., perpendicular to the tube surface ), but they also cling to one another and agglomerate. In this configuration, the steel rods push ed the magnet ic abrasive against the tube inner s urface, and t he finished surface consisted of the accumulation of deep scratch es. In the case of stainless steel rods (Figure 3 14 ( B )) the axes of the magnetized rods are generally aligned with the magnetic flux but th ere is repulsion between the individual rod chain s, so a differently configured mass pushed the magnetic abrasive against the tube surface. When subjected to the unevenness of the tube surface, the rods show instability, e.g. relocat ing and impacting the surface because of the magnetic anisotropy and reduced magnetic force. The magnetic abrasive was pushed by unstable rods and showed discontinuous contacts against the tube surface. This enhanced the material removal but hardly result ed in continuous surfac e finishing. In contrast, the interior volume of capillary tubes is limited, so the rods must be introduced into the tube with their axes parallel to the tube axis Consequently, the rods in capillary finishing behave differently from those in the cases o f large sized tube finishing. M o reover, the rods may help to keep the flexible capillary tube straight at the
49 finishing area, allowing a flexible tube to be treated as a straight tube. Before applying the ferrous rods to the finishing experiments, the moti on of the rods was observed using a transparent nonferrous polymer tube. T o facilitate the observation large sized iron particles (150 300 m ) were introduced in place of magnetic abrasive. Figure 3 14 P hotographs of c arbon steel rod and 304 stainless steel rod inside nonferrous polymer tube (Photograph courtesy of Junmo Kang) Figure 3 14 shows photographs of carbon steel (0.80 0.90 wt% C) and stainless steel rods and iron particles inside nonferrous polymer tube while the tube is rotated at 200 min 1 t o facilitate visual observation. Tube rotation is measured by tachometer. The magnetic flux flow s from one pole tip to another, and the ferrous rod and iron particles are magnetized following the magnetic flux. In the case of the carbon steel rod, magnetic force causes the iron particles to gather as a mass around the rod ( Figure 3
50 14( A ) ) due to the ferromagnetism of both the carbon steel rod and the iron particles. The mass pushes the inner surface of the rotating plastic tube with strong magnetic force a nd, w hen the pole tips are reciprocat ed in the axial direction, the mass follows the pole tips stably and produce s scratches on the tube surface In the case of the stainless steel rod ( Figure 3 14( B ) ) the iron particles are attracted by the magnetic fie ld more than the rod is, and in the restricted area inside the tube the stainless steel rod settles into a position over the iron particles while slightly leaning toward the line of magnetic force. A t the contact area between the iron particles and the r od, there is no interaction by magnetic force between them. Only one end of the rod pushes the iron particles against the tube surface and participates in the finishing action. As the carbon steel rod does, the stainless steel rod ( with the iron particles ) follows the motion of the pole tips when the pole tips are fed in the axial direction. However, the weak magnetic force acting on the rod causes unstable conditions and t he rod occasionally shows rotation about its axis over the iron particles. When the pole tips change direction during reciprocation, the stainless steel rod does not immediately change direction to follow the pole tip motion in the same way that the carbon steel rod does. Because of the magnetic anisotropy and low magnetic susceptibility the stainless steel rod requires a strong magnetic field to be attracted to the pole tips. Regardless of the pole motion, the rod is stationary until the magnetic field becomes strong enough to attract the rod, and it was observed that one end of the rod needs to be close to the pole tip edges. After the rod is attracted by the field, it follows the motion of pole tip s As a result, the reciprocating stroke of the rod is shorter than the stroke of the pole tip reciprocation Occasionally, the rod moves so that the other end of
51 the rod push es the iron particles against the inner surface of the tube. Considering th ese effects the finishing area for the stainless steel rod is typically shorter than the area when using the carbon steel rod. I n Section 3.3.1 the effects of the tool behavior ( due to the geometric and magnetic properties ) on the internal finishing characteristics will be experimentally studied with straight capillary tubes. Iron particles and rods of carbon steel and stainless steel are prepared for the tests. 3.3.1 Experimental Setup and Conditions Table 3 2 shows the experimental conditions, for this study, 304 stainless steel tubes ( straight, OD 0.64 mm, ID 0.48 mm, length 100 mm initial surface average roughness ~ 1 m Ra ) and flexible capill ary (Tube B) were used as workpieces. In each test, the tube was chucked 12.5 mm from one end. Centrifugal force acts on the capillary tube when it is rotated at high speed, and to diminish the run out of the tube beyond the finishing area, the opposite en d of tube was held by a flexible jig, which can accommodate changes in the tube length during the rotation T he tube rotational speed was set at 2500 min 1 T he reciprocating stroke of the 4 mm wid e pole was set at 4 mm in the axial direction and the pol e feed rate was set at 0.6 9 mm/s, which has been found to be appropriate for deburring. Because of the small tube diameter, the stylus of a surface roughness profilometer, the device most commonly used for surface roughness measurement, could not be insert ed inside the capillary. This created difficulty in measuring and tracking the changes in the surface roughness and burr heights with finishing time. Instead, t he finishing tests were performed for a certain period, then the tube was cleaned using ethanol in an ultrasonic cleaner, sectioned along the tube axis and t he
52 inner surface was evaluated using an optical profilometer and microscope T he changes in the material removal were tracked by measuring the weight with a micro balance (10 g resolution ). Tab le 3 2 Experimental conditions of internal finishing of straight capillary tube Workpiece Straight tube: 304 stainless steel tube ( 0.64 0.48 100 mm) Workpiece revolution 2500 min 1 Magnetic abrasive Aluminum oxide (WA) particles (80 m mean dia.) in magnetic abrasive (<10 m) Magnetic tool Iron particles ( 50/+100 mesh, 150 300 m diameter ) Carbon steel rod ( 0.25 4 mm), 304 Stainless steel rod ( 0.24 4 mm) Lubricant Soluble type barrel finishing c ompound Permanent magnet Nd Fe B permanent magnet : 10 12 18 mm Pole reciprocating motion Amplitude: 4 mm, Speed: 0.6 9 mm/s Workpiece Pole clearance 0.11 mm Pole tip T able 3 1 3.3.2 Effects of Material Properties of Magnetic Tools Initially the relationship between the finishing time and surface ro ughness improvement was evaluated to determine the duration of experiments using mixed type magnetic abrasive (a mixture of 80 m mean diameter magnetic abrasive and iron particles in the 150 300 m range which was the largest range that could be introdu ce d into the tube without clogging ) [ 52 ]. To determin e the appropriate amount of the mixed type magnetic abrasive, the volume of the mixture wa s compared to the total volume inside a length of tube equal to the width of the pole tip. A mixture of 0.13 mg m agnetic abrasive and 0.49 mg iron particles which results in 41.1 vol% occupied by the mixed type magnetic abrasive was supplied for the experiments. After 117 cycles (22.5 min), the roughness improvement slowed down and this became the index used for the
53 subsequent experiments. The finishing experiments were repeated under each set of condition s at least three times to confirm the repeatability of the results Figure 3 15 P hotographs and surface roughness profiles of tube interior before finishing, s urface finished with iron particles s urface finished with carbon steel rod, and s urface finished with stainless steel rod for 22.5 min Figures 3 15 show ( A ) photographs of tube interiors and ( B ) representative surface roughness profiles measured by an o ptical profiler. Figure 3 16 shows the three dimensional surface of F igure 3 15. With a mixture of magnetic abrasive and iron particles, the mass of the mixture nearly followed the motion of the pole tips, and the finished length was 5 6 mm, although the p ole tips covered 8 mm (4 mm pole tip width and 4 mm stroke) in the tube axis direction. This was a result of the friction between the mass and tube surface, which caused a delay in the following motion of the abrasive The chains of iron particles and magn etic abrasive conform to the shape of the tube interior This enables uniform internal finishing of capillary tubes. W ithin the finished area t he surface was almost uniformly finished from 0.75 m Ra to 0.03 m Ra ( Figure 3 15( A ) and ( B ) )
54 When a carbon s teel rod was inserted with the magnetic abrasive the finished length of the surface was about 7 mm. The steel rod and magnetic abrasive (0.25 mg) occupied 43.7 vol% at the finishing area. The rod and magnetic abrasive were both strongly attracted by the p ole tips and more closely followed the pole motion. The representative finished surface s are shown in Figure 3 15( C ), and a representative surface roughness value was 0.27 m Ra W ith strong magnetic force t he steel rod pushed the magnetic abrasive sandwi ched between it and the tube surface. Depending on the configuration of the magnetic abrasive pushed by the rod, the magnetic abrasive irregularly generate d deep scratches on the tube surface. This resulted in the larger surface roughness than that in the case with iron particles (no rod) Moreover, t he material removal with the steel rod (0.18 mg) was nearly twice as high as the case with iron particles (0.1 mg). In the conditions with the stainless steel rod, the material removal (0.02 mg) was not enough to completely remove the initial surface texture. The magnetic abrasive (0.25 mg) and rod occupied 41.6 vol% at the finishing area ; however, due to the tilted position of the rod, only one end of the rod participate d in the finishing action. Fig ure s 3 15 ( d ) show that the initial texture partially remained on the finished surface, and the roughness was measured to be 0.39 m Ra The finished length was measured to be only about 1.5 2 mm along the tube axis. T his is attributed to the irregular rod motion due to the weak magnetic force and the anisotropic magnetic properties. Accordingly, the effects of the ferrous tools on the finishing characteristics can be summarized : (1) The chains of iron particles and magnetic abrasive conform to the shape of the tube interior This enables uniform internal finishing of capillary tubes. This
55 method is sufficient for operations that require small material removal, such as tube surface finishing (2) The carbon steel rod pushes the magnetic abrasive against the tube surfa ce with strong magnetic force, generating high material removal. This method is appropriate for conditions that require large material removal ( such as deburring ) but not for surface finishing, and (3) T he stainless steel rod generates weak magnetic force with its magnetic anisotropy. T his results in the slow material removal rate and relatively short er finished area. This method is insufficient for performing either surface finishing or deburring in a timely manner. Figure 3 16 Three dimensional surface shapes of tube interior before finishing, s urface finished with iron particles s urface finished with carbon steel rod, and s urface finished with stainless steel rod for 22.5 min
56 Figure 3 17 P hotograph and roughness profile of s urface finished with sta inless steel rod, magnetic abrasive, and iron particles For the internal finishing of flexible capillary tubes, i t is necessary to introduce a rod inside the capillary tube to make it straight for processing. While t he carbon steel rod is a promising ferr ous tool for deburring purpose s it is not appropriate for surface finishing purpose s If the material removal rate using the stainless steel tool rod could be improved, this method may be another potential method for both deburring and surface finishing. The lack of relative motion between the tube surface and the magnetic abrasive pushed by the stainless steel rod can be overcome by extending the stroke of the rod reciprocati on For example, increasing the stroke of the pole tip from 4 to 8 mm result ed in a finished length of about 11 mm instead of 12 mm. Furthermore a part of magnetic abrasive can be replaced with iron particles, since some of the magnetic abrasive did not directly participate in the mater i al removal The mixed type magnetic abrasive wil l increase the magnetic force. Accordingly, a portion of the magnetic abrasive (65 wt%) was replaced by iron particles, and the finishing experiments were performed for 51.3 m in (the pole tips were reciprocated for 8 mm in 117 times at 0.6 mm/s)
57 Figure 3 17 sh ows surface roughness profiles of the finished surface as measured by the optical profilometer. The material removal increased from a previous value of 0.02 mg to 0.10 mg and the surface consist ed of an accumulation of almost evenly generated scratc hes ( the roughness value was 0.09 m Ra ). The micro asperities left from the initial surface must be removed by the extending the finishing period In the hybrid conditions (stainless steel rod and iron particles), t he role of the rod is to drive both the iron particles and magnetic abrasive in the axial direction and the role of the iron particles is to form the mixed type magnetic abrasive, increase the magnetic force acting on the magnetic abrasive, and encourage the material removal. This hybrid method is applicable for simultaneous surface finishing and deburring. In magnetic abrasive finishing magnetic force acting on magnetic abrasive determines the finishing force as well as the magnetic abrasive behaviors. To clarify the finishing mechanism of ca pillary tube finishing using the magnetic tools, the measurement of magnetic for ce acting on magnetic abrasive wa s measure d The system for magnetic force measurement using strain gauge has been designed (Figure 3 1 8 ) and implemented. Two s train gages (Ha lf bridge method, gage factor: 2.09 1.0%, gage resistance ( 24 C, 50%RH): 119.8 0.2 ) we re attach ed on the aluminum plate (14.6x3.11x150 mm ) and a magnetic tool with particles wa s located in the groove on the plate for preventing the dispersion of particles The plate located on the 3 axis micrometer stage I n order to avoid the contact between the bottom of plate and the top of pole tip, the clearance between the plate and pole tip was set at 0.1 mm Each measurement wa s repeated three times, and the standard error s were indicated on the ea ch bar in Figure 3 19 and Figure 3 20
58 Figure 3 18 Schematic of magnetic force measurement system for single pole tip system and photograph for top view of plate (Photograph courtesy of Junmo Kang) Each ferrous particles (carbons steel tool, 304 stain less steel tool, iron particle and magnetic abrasives) are prepared in same volume percent (21 vol%) and the magnetic force are measured. The iron particle has twice magnetic force than that of magnetic abrasive (Figure 3 19), and this has been shown in pr evious study [50 ]. W ithin 21 volume percent of ferrous tools in flexible capillary tube (Tube B), carbon steel tool ( 0.5 4 mm, 6.71 mg) is 3.7 times stronger than the iron particles (150 300 m diameter,1.5 mg) and has 2.5 times stronger force than 304 st ainless steel tool itself ( 0.51 4 mm, 6.75 mg). Using the 304 stainless steel tool instead of iron particle shows 50 percent incensement of magnetic force. This helps the magnetic abrasive follow the pole along the tube axis and enhances the material remo val rate during the process. Only mixed type magnetic abrasive could not achieve the internal finishing of
59 flexible capillary tube due to its both small particle size and less magnetic force. The magnetic tools (carbon steel and 304 stainless steel rods) c an increase the magnetic force acting on magnetic abrasive. Moreover the substitution of magnetic abrasive to iron particles assists the flexible chain forms on the surface strongly. Figure 3 19 Relationship between finishing force and magnetic tool in same volume percent The case of carbon steel tool ( 0.25 4 mm, 1.54 mg) and magnetic abrasive (0.25 mg) show ed the strongest magnetic force due to its high susceptibility. This made the highest material removal as well as deep scratc hes on the surface. Th is force wa s 1.6 times stronger than the mixed type magnetic abrasive (0.6 mg) and 2.6 times stronger than that of 304 stainless steel case. The case of 304 stainless steel tool ( 0.24 4 mm, 1.55 mg) and magnetic abrasive (0.25 mg) exhibit ed the weakest ma gnetic force as expected due to its magnetic properties With the tool instable motion during process, this case resulted in insufficient material removal. In order to increase the magnetic force, some portion s of magnetic abrasive were substituted with ir on
60 particle (hybrid method). The mixture of 304 stainless steel tool, iron particle (0.24 mg), and magnetic abrasive (0.06 mg) were applied and show ed stronger magnetic force than the mixed type magnetic abrasive case. Even though the force of hybrid metho d is weaker than the case of carbons steel tool, the method produces smoother surface with uniform surface asperities as well as with increased material removal. This can allow the method is applicable for the internal deburring of flexible capillary tubes Figure 3 19 Relationship between finishing force and experimental condition 3.3.3 Hybrid Finishing Methods for Flexible Capillary Tube Based on the fundamental understanding of the ferrous tools, the hybrid method was applied to the internal finishin g of flexible capillary tubes. Experiments were conducted using 304 stainless steel flexible capillary tubes (Tube B: OD 1.36 mm, ID 1.02 mm, Figure 3 1( B )). D uring the micro laser machining process, molten material re solidified and adhered all over the s urface ; burr heights were measured by an optical
61 profilometer to be in a range between 70 and 90 m. The slit pitch (distance between slits) was about 400 m, as shown in Figure 3 2 1 ( A ). At first, mixed type magnetic abrasive, a mixture of 1.20 mg iron par ticles and 0.33 mg magnetic abrasive, was inserted into the tube with a carbon steel rod ( 0.50 4 mm) The abrasive mixture and the rod occupied 4 6.8 vol% of the finishing area of the tube. The finishing experiments were performed for 30 min, during which time the poles made a total of 155 four millimeter reciprocating stroke s at 0.69 mm/s As shown in Figure 3 2 1 ( B ), the r e solidified material adhered to the surface and edges w ere clearly removed after finishing. The surface was not finely finished (0.24 m Ra 0.2 mg/mm ), which was previously shown to be a trend of the process using a carbon steel rod. Table 3 3 Experimental conditions of internal deburring of flexible capillary tube with laser machined slits: Tube B Workpiece Tube B: 304 stainless ste el tube ( 1.36 1 02 100 mm) Workpiece revolution 2500 min 1 Magnetic abrasive Aluminum oxide (WA) particles (80 m mean dia.) in magnetic abrasive (<10 m) Magnetic tool Iron particles ( 50/+100 mesh, 150 300 m diameter ) Carbon steel rod ( 0.5 4 mm) 304 Stainless steel rod ( 0.51 4 mm) Lubricant Soluble type barrel finishing c ompound Permanent magnet Nd Fe B permanent magnet : 10 12 18 mm Pole reciprocating motion Amplitude: 4 mm, Speed: 0.6 9 mm/s Workpiece Pole clearance 0.11 mm Pole tip Tabl e 3 1 The experiments were also performed with stainless steel rod ( 0.51 4 mm). The finished edge and surface re t ain ed the burrs and re solidified materials ( 0. 19 m Ra 0.11 mg/mm) in Figure 3 21( C ) The pole reciprocati on stroke was extended from 4
62 to 8 mm, and the pole reciprocation was set for 312 strokes in 120 min. The abrasive mixture 1.25 mg iron particles and 0.36 mg magnetic abrasive and the rod occupied 4 8.8 vol% of the finishing area. Lubricant was added after every 78 pole strokes. As show n in Figure 3 2 1 ( D ), the burrs were completely removed, and the inner surface was finished to 0.05 m Ra with the rate of 0.27 mg/mm. Figure 3 2 1 P hotographs of tube interior before and after finishing; As received surface, s urface finished with carbon steel rod, magnetic abrasive and iron particles for 30 min s urface finished with stainless steel rod, magnetic abrasive, and iron particles for 3 0 min and s urface finished with stainless steel rod, magnetic abrasive, and iron particles for 12 0 min In the carbon steel tool, s trong magnetic force acting on magnetic abrasive allows removing the burrs completely in short period with high material removal. However, the force caused by 304 stainless steel tool, even though the magnetic force is stronger than that of iron particles, still shows insufficient material removal and requires more processing time for successful surface and edge finishing of flexible capillary tube
63 CHAPTER 4 MULTIPLE POLE TIP SYSTEM USING A M ETASTABLE AUSTENITIC STAINLESS STEEL TOO L 4.1 Introduction of Multiple Pole tip System U s e of a magnetic tool facilitates finishing efficiency (reduces processing time) by control of magnetic force. The roles of magnetic tool are to increase magnetic force and to straighten the tube. However, i t still requires long finishing time due to limitation in the numbers of finishing spot. To resolve this problem, it is desired to finish multiple areas simultaneously. In order to satisfy this demand, a new method which uses a metastable austenitic stainl ess steel tool has been proposed. A metastable austenitic stainless steel rod alternating magnetic and nonmagnetic regions in a body can be fabricated t hrough selective heat treatment. Magnetic abrasive is attracted to the borders of the magnetic regions o f the developed tool to create additional finishing points when it magnetized In combination with a multiple pole tip system, this unique magnetic property facilitates simultaneous finishing of multiple regions for shortening finishing time. C hapter 4 describes the fabrication, the crystalline structure, and the resulting magnetic properties of the heat treated metastable austenitic stainless steel tool. T he magnetic abrasive behavior, the finishing characteristics, and a mechanism to extend the finish ed length are clarified for internal finishing of capillaries. A new system is proposed to improve the finishing efficiency by increasing the number of finishing points and by short ening the length of the pole stroke, and it is used in combination with a p artially heat treated stainless steel tool. This system is called a multiple pole tip system.
64 A S ingle pole tip system B M ultiple pole tip system Figure 4 1 Schematics of single pole tip system and multiple pole tip system Figure 4 1( A ) shows a schematic of a typical magnetic abrasive finishing setup with a single pair of pole tip which called as a single pole tip system The desired magnetic field at the finishing area is generated by t wo permanent magnets attached to a steel yoke. A mixed typ e magnetic abrasive introduced into the tube is pushe d by magnetic force against the tube surface and finishes the tube w h en the tube is rotated at high speed. The pole tip width is the default finished length, and th e accumulation of multiple short length finishing passes is needed for long tube finishing.
65 Figure 4 1( B ) conceptually shows the proposed method to extend the default finished length. An additional pair of pole tips is added (yoked together as shown Fig. 4 1( B ) ) and a long magnetic tool is plac ed inside the tube. T he mixed type magnetic abrasive is introduced into two sections inside the tube; this doubles the default finished length and, thus, doubles the finishing efficiency A tool with alternating magnetic and non magnetic sections is introd uced inside the tube with a mixture of magnetic abrasive and iron particles The length of the magnetic section of the tool corresponds to the pole tip width. T he magnetic abrasive follow s the lines of magnetic forc e and accumulates at the border s of the m agnetic sections of the tool, thereby creating multiple finishing areas. The magnetic abrasive pushes the inner surface of the tube and, when the tube is rotated at high speed, it exhibits relative motion against the tube surface and removes material. By feeding the magnetic pole tip assembly along the tube axis, the magnetic abrasive and the tool are both dragged by magnetic force and the finished area is extended 4.2 Magnetic Properties of Metastable Austenitic Stainless Steel Tool The key to realizin g the multiple pole tip system is the use of a tool that has alternating magnetic and non magnetic sections; this is accomplished by using a metastable austenitic stainle ss steel tool, 304 stainless steel is used for this study Once the stainless steel ha s undergone cold working or strain hardening by plastic deformation it experiences a martensitic transformation and exhibits ferromagnetism. However, the austenitic phase can be retrieved (and thus exhibit nonmagnetic properties) by heat treatment beyond the Curie temperature (at least 600 C) [ 6 0 61 ] This treatment can make multiple alternations in the magnetic property of a single tool. In this study, the tool was partially heat treated using the flame of butane lighter in
66 ambient condition s for 30 s. A fter the heat treatment, the tool was cooled in air to room temperature Figure 4 2 Tool geometry and photograph of p artially heat treated stainless steel tool with iron particles (Photograph courtesy of Junmo Kang) Figure 4 2 shows the geometry and a p hotograph of partially heat treated stainless steel tool with iron particles The untreated sections of the tool exhibit magnetic anisotropy due to the plastic deformation of the manufacturing process, and t he heat treated section exhibits paramagnetism du e to fully austenitic condition The iron particles are attracted to the borders of the ferromagnetic sections T he crystal structures of the tool were characterized using an X ray diffractometer (XRD: Copper : 1.54 )) at room temperature. The X ray beam penetrates 5 m from the surface Figure 4 3 shows XRD patterns of the untreated and the heat treated section s o f the stainless steel tool. B oth body center ed cubic (bcc) and face centered cubic (fcc) structure s are observed in the untreated section. The bcc structure must have been generated during the previous tube drawing p roce ss In the heat treated section onl y an fcc structure is observed. This confirms that the applied treatment locally retrieved the fcc structure (austenitic phase) in t he surface layer.
67 Figure 4 3 Partially heat treated stainless steel tool and X ray diffraction patterns ; Untreated sectio n and Heat treated section 4.3 Finishing Characteristics in Multiple Pole tip System 4.3.1 A Method to Deliver Magnetic Abrasive Deeper into Capillary Tube For capillary tube finishing, the finished surface quality obtained using the stainless steel tool has not reached the level of the single pole tip (pair) system [3 5 ]. Moreover, it was observed that unstable tool motion occasionally creates a poorly
68 finished surface. Practically, the finishing area corresponding to the pole tip farthest from the open e nd of the tube (i.e., toward the machine chuck) tends to show a poorly finished surface. U sing magnet ic pole tips placed outside the tube magnetic abrasive can be introduced into the tube by means of magnetic force ; however, some magnetic abrasive typical ly remains on the inner surface of the tube instead of being dragged by the magnet. This is because of friction against the tube inner surface and is unavoidable. T o alleviate this situation the magnetic force acting on the magnetic abrasive must be incre ased e nough to overcome the friction. The force can be controlled by the magnetic field, and adjusting the pole tip geometry is a simple way to control the magnetic field [ 6 5 ]. Two kinds of pole tip s were prepared ( as shown in Figure 4 4( A ) ) to examine the effects of the magnetic field on the delivery of the magnetic abrasive into the area corresponding to the chuck end pole tip : straight and tapered in the axial direction Figure 4 4( B ) shows the relationship between the magnetic flux density ( measured by a hall sensor ( 1.0 mm sensing area) ) of the pole tip set and the distance X The condition in which the center of the sensing area is placed over the pole edge was X =0 mm. A higher magnetic flux density and a larger gradient were obtained above the tapere d pole tip B compared to the values measured above the straight pole tip A [6 6 ] This indicates that the tapered pole tip B generate s greater magnetic force acting on the magnetic abrasive. The interaction of the pole tips increased the magnetic flux dens ity between the pole tips. The magnetic abrasive tends to be attracted by these inward edges more than outward edges.
69 Figure 4 4 G eometr y of pole tip set geometry and pole tip shapes and changes in magnetic flux density Photograph of finishing unit is shown in Figure 4 5 and t he finishing conditions are listed in Table 4 1. The case where pole tip A is mounted close to the free end of the tube is called the 18 8 9 pole tip set, and the opposite configuration is called the 9 8 18 pole tip set. In the bot h cases, the two pole tip sets were mounted at an angle of 90 as illustrated in Figure 4 1( B ). T he supplied amount of magnetic abrasive was determined based on the space inside the tube corresponding to the pole tip, and 47 % of the volume was occ upied b y the magnetic abrasive.
70 Figure 4 5 External view of multiple pole tip system (9 8 18 pole tip set) (Photograph courtesy of Junmo Kang) As Table 4 1 show s 8 mg for pole tip A and 4 mg for pole tip B were initially supplied in to the tube. When the pole tip is fed toward the chuck end of the tube, the magnetic abrasive suspended at the area corresponding to the other end of pole tip A moves toward the chuck end but some of the magnetic abrasive particles adhere to the surface o f the tube due to friction. When the pole tip return s to its original position, the chuck end pole tip attracts some magnetic abrasive and drags it along the tube surface. This back and forth motion can be used as a mechanism to convey the magnetic abrasiv e in to area s deeper in the tube. The pole stroke was initially set at 26 mm, which is the length that causes the inner edge of pole tip B to reach the farthest edge of pole tip A in the 9 8 18 case
71 Table 4 1. Experimental conditions for 9 8 18 and 18 8 9 pole tip set Workpiece 304 stainless steel tube ( 1.27 1 06 100 mm) Pole tip type 9 8 18 pole tip set (18 mm inside) 18 8 9 pole tip set (9 mm inside) Workpiece revolution 2500 min 1 Ferrous particle Iron particles ( 50/+100 mesh, 150 300 m dia .): 80 wt% + Aluminum oxide (WA) particles (80 m mean dia.) in magnetic abrasive (<10 m): 20 wt% Amount of ferrous particle Pole tip A: 8 mg, Pole tip B: 4 mg Magnetic tool Heat treated 304 Stainless steel rod: 0.51 35 mm Lubricant Soluble type barr el finishing c ompound (pH 9.5, 755 mPas at 30C) Pole reciprocating motion Speed: 0. 59 mm/s Stroke length: 26 mm Number of strokes: 117 Workpiece Pole clearance 0.11 mm Processing time 174 min Figure 4 6 Changes in surface roughness with distanc e X Chuck Pole tips A B Tube Chuck Tube A B Pole tips
72 Figure 4 6 shows the changes in surface roughness Ra with distance X While the 18 8 9 pole tip set produced a uniformly finished surface, the roughness increased with the distance X in the case of the 9 8 18 pole tip set. In the 18 8 9 case, the magne tic abrasive is sufficiently distributed by the magnetic field to produce a uniformly finished surface. In the 9 8 18 case, the area corresponding to the chuck end edge of pole tip A must have the least amount of magnetic abrasive before any pole tip feed motion The area beyond X =17 mm i n the 9 8 18 condition must lack magnetic abrasive from the beginning of the experiments and t he area between X =17 and X =43 mm must be finished using the magnetic abrasive delivered by the abovementioned mechanism However the amount of the magnetic abrasive decreases with increasing distance X and must be insufficient for finishing. T he experiments confirm that the magnetic abrasive must be sufficient ly distribut ed and that t he magnetic field distribution and pole feed le ngth are key s to effectively convey the magnetic abrasive. 4.3.2 Effects of Heat treated Sections on Internal Deburring of Flexible Capillary Tubes with Multiple Laser machined Slits Experiments were conducted using 304 stainless steel flexible capillary t ube (1.36 mm OD, 1.02 mm ID, and 100 mm long). R andom irregular burrs and obstacles 50 ~ 70 m in h e igh t, can be seen in Figure 4 7( A ) which shows representative micrographs of the as receive d surface. The sl o t pitch (distance between sl o ts) was about 400 In addition to the pole stroke length of 26 mm, a short stroke length of 8 mm was applied for this study. The gap between pole tips is 8 mm, so that is theoretically the shortest stroke that can convey the magnetic abrasive from the inner edge of pole tip A to the adjacent edge of pole tip B. The other conditions are shown in Table 4 1.
73 In the case of the 8 mm pole stroke, burrs and obstacles remained in the area correspond ing to pole tip B ( beyond X =24 mm which is slightly shorter than the pole tip w i d th and pole stroke length combined: 18+8 mm ) The burrs and obstacles must prevent the introduction of the magnetic abrasive into the area, and the magnetic abrasive must stay at the area corresponding to pole tip A. While the 8 mm stroke causes the inne r edge of pole tip A to reach the adjacent edge of pole tip B the lack of any overlap of these edges makes the stroke too short to adequately convey the abrasive to pole tip B. As a result, the lack of mixed type magnetic abrasive beyond that point allows burrs and obstacles to remain Figure 4 7 Intensity maps of surface and three dimensional surface shapes measured by optical profiler On the other hand, the overlap of the poles with the 26 mm pole stroke must deliver the abrasive more completely (than the 8 mm stroke) to the area corresponding to pole tip B, resulting in the successful machining Figures 4 7( A ) and ( B ) show the
74 intensity maps and three dimensional surface shapes ( measured by optical profiler ) of the as received surface and surface fini shed with the 26 mm pole stroke length conditions respectively It is seen that both burrs and obstacles were removed by the proposed method and that the surface was uniformly finished from 2.5 3.5 m Ra to 0.3 0.4 m R a. This demonstrat es that the propose d abrasive delivery method enables the feasibility of the MAF process for the internal finishing of flexible tubes, regardless of the presence of large burrs and obstacles. 4.3.3 Effects of Heat treated Sections on Finished Surface and Abrasive Behavior T he effects of t he tool behavior in the multiple pole tip system were experimentally studied using tools with three kinds of magnetic properties. Figure 4 8 shows a photograph of four magnetized 54 mm long tools with iron particles : Tool A is made of carbon steel Tool B is 304 stainless steel and Tool C is partially heat treated 304 stainless steel In the case of Tool A the iron particles are mo re attracted by the tool ends because of the re sidual magneti sm and magnetic shape anisotropy Tool B exhibits magnetic anisotropy, which was generated during cold work ( deformation induced martensite transformation ), and the iron particles are attracted to the tool ends only. In the case of Tool C t he iron particles are attracted to both ends of magnetic regions following the residual magnetism T he center region (18 mm long) of the stainless steel Tool C was heat treated, and it has a f ace centered cubic structure [3 5 ]. This crystal transformation diminishes the magnetism and divides the magnetic section into two regions [3 5 62 63 ].
75 Figure 4 8 Magnetized tool s A, B, C and D with iron particles (Photograph courtesy of Junmo Kang) Figure 4 8( D ) shows the configuration of the heat treated magnetic tool ( Tool D ) with a photograph of magnetic abrasive with the tool based on that design. The magnetic abrasive is attracted to the border s of the magnetic region s. Compared to Tool C (Fig ure 4 8 ( C )), this doubled the number of the places where the magnetic abrasive is held Each tool was inserted in a clear polypropylen e plastic tube (3 mm outer diameter, 0.7 mm wall thickness), which was set on the machine shown in Figure 4 10. Eight milligrams ( 8 mg ) of a mixture of iron particles and magnetic abrasive was supplied with the magnetic tool. The pole tip set was fed at 0. 59 mm/s, and the feed length was set at 18 mm. T h e tube was rotated at 200 min 1 and the behavior of the tool and mixed type magnetic abrasive inside the tube was observed.
76 Figure 4 9 Motion s of each tool with mixed type magnetic abrasive for a pole t ip stroke length of 18 mm Fig ure 4 9 schematic ally shows the behavior of the magnetic tool s with mixed type magnetic abrasive while the tube is rotated Figure 4 9 ( A ) is the case with Tool A Initially, the mixed type magnetic abrasive was supplied into tw o regions corresponding to the pole tips. Once the pole tip set started moving, some mixed type magnetic
77 abrasive stuck to the carbon steel tool surface regardless of the pole tip motion because of the ferromagnetism of the tool. The carbon steel tool push e s the mixed type magnetic abrasive against the inner tube surface with a strong magnetic force. This increases the friction force between the tool and inner surface of the tube and this friction force p revented the tool from following the pole tip feed i n the experiment. Figure 4 9( B ) shows the case of Tool B which exhibits anisotropic magnetism and lower magnetic susceptibility than t he mixed type magnetic abrasive The mixed type magnetic abrasive separated into two finishing sections because it i s at tracted to the pole tips more than to Tool B Tool B simply l ies on the mass of magnetic abrasive. While the pole tips are moving the mixed type magnetic abrasive follows the pole tip motion Both ends of Tool B are attracted to the magnetic field accordi ng to its anisotropic magnetism ; Tool B thereby stay s in its initial position regardless of the pole tip and magnetic abrasive motion Fig ure 4 9 ( C ) shows the case with Tool C. The borders of magnetic regions of the tool correspond to the pole tip edges a nd are attracted to the magnetic field. The mixed type magnetic abrasive is attracted by the pole tips and once the pole tips are f ed along the tube axis; both the mixed type magnetic abrasive s and tool follow the pole tip motion As a result, t he magneti c abrasive shows smooth relative motion against the tube inner surface needed for internal tube surface finishing. This experimental observation demonstrates the conditions required facilitating the deliverability of the tool and magnetic abrasive in the m ultiple pole tip system by conforming to the pole tip motion, that is, the tool must have alternating magnetic and
78 non magnetic regions, and the borders of the magnetic region of tool must co rrespond to the pole tip edges. In the multiple pole tip system u sing the 54 mm long Tool C, the default finished length is the region corresponding to two 18 mm wide pole tip s. If the pole tip feed is sufficient to cover the gap between two pole tips (18 mm), the 72 mm length should be finished uniformly. In practice, some mixed type magnetic abrasive adheres to the surface of the tube because the friction between the mixed type magnetic abrasive and tube surface exceeds the magnetic force acting on the mixed type magnetic abrasive to follow the pole tip motion. The reg ion corresponding to the chuck end edge of the pole tip must have the least amount of magnetic abrasive As a result, the deeper the finishing area was in the tube, the rougher the finished surface was due to the lack of magnetic abrasive. Accordingly, the initial insertion of magnetic abrasive deeper into the tube and the deliverability of the mixed type magnetic abrasive play important roles in accomplish ing the desired finishing performance. The alternating magnetic property and the interval between the magnetic and nonmagnetic regions must be designed to satisfy these matters. In the case of Tool C ( Figure 4 9 ( C )), magnetic flux flow s from one end of the magnetic region to the other end, and the magnetic abrasive is attracted following the flow of magnet ic flux. In turn, tube surface finishing is pre dominantly perform ed in four places. If the length of the magnetic region is reduced from 18 mm to 3 4 mm and a region is created on the tool corresponding to each pole tip edge, the magnetic flux concentrate s at the pole tip edges, doubl ing the number of borders to attract magnetic abrasive. F inishing is dominantly performed at the four regions where the mixed type
79 magnetic abrasive is encouraged to remain by the tool/pole configuration Visual observation of Tool D with mixed type magnetic abrasive in a transparent polymeric tube ( Figure 4 9 ( D )) showed that Tool D holds the mixed type magnetic abrasive at four places and that the pole tip motion is smoothly follow ed The geometry of pole tip sets in both sing le and multiple pole tip system s used in experiment are shown in Fig ure 4 10 The width of the pole tip, which defines the default finished length, is 18 mm. The pole tip set for the single pole tip system has two magnets. In the case of the multiple pole tip system, each pole tip has one magnet, and the two magnets are coupled by a steel yoke which is 54 mm long ( parallel to the workpiece axis ) Figure 4 10 G eometr y of single and multiple pole tip sets Figure 4 11 shows changes in magnetic flux density B y measured by a Hall sensor (sensing area: 1.0 mm) with distance X in both pole tip systems. There is no significant difference in the magnetic flux density B y between the two systems. Both systems show a similar trend : the magnetic flux density and i ts gradient increase from the center toward the edges of pole tip s T he magnetic abrasive is attracted by magnetic f orce toward the pole tip edges. W ithout assistance of a magnetic tool this
80 trend encourages the plugging of the magnetic abrasive in the re gion corresponding to the pole tip edges w hen the tube ID is less than 1 mm. Table 4 2 shows the experimental conditions. Stainless steel tubes (1.27 mm OD, 1.06 mm ID, 100 mm long, initial surface roughness: 2 Rz ) were prepared Ten millimeters at o ne end of the tube wa s chucked, and the other end was supported by a flexible jig to reduce run out during rotation. The tube rotational speed was set at 2500 min 1 T h e pole tip feed length and rate w ere set at 18 mm and 0.59 mm/s, respectively. The finis h ing experiments began 5 mm from the free tube end In the case of the single pole tip system, mixed type magnetic abrasive 15 mg (80 wt% ferrous particles and 20 wt% magnetic abrasive) was supplied, which took up 45.3 % of the enclosed volume of the finis hing area. In the multiple pole tip system, 47.3 % of the volume at each finishing area was taken up (24.2 % by the mixed type magnetic abrasive and 23.1 % by the stainless steel tool). In supplied into the tube after every 20 pole tip strokes For each finishing experiment, the total number of pole tip strokes was 180. After each experiment, the tube was cleaned using ethanol in an ultrasonic cleaner and sectioned along the tube axis. The in ner surface was evaluated using an optical surface profiler and a stereomicroscope The optical surface profiler has a lateral resolution of 275.7 nm and a vertical resolution of <0.1 nm. The material removal was calculated by comparing the weight (measure d using a micro before and after finishing. The finishing experiment was repeated at least three times under each condition to confirm the representative trends using the finishing equipment shown in Figure 4 12.
81 Figure 4 11 C hanges in magnetic flux density of single and multiple pole tip system with distance X Table 4 2. Experimental conditions for 18 18 18 pole tip set Workpiece 304 stainless steel tube ( 1.27 1.06 100 mm) Workpiece revoluti on 2500 min 1 Pole tip feed Speed: 0.59 mm/s, Number of strokes: 180 Workpiece pole tip clearance 0.3 m m (Polytetrafluoroethylene (PTFE) tape thickness) Lubricant Soluble type barrel finishing compound (pH : 9.5, Viscosity: 755 mPas at 30C) Pole tip system Single pole tip system Multiple pole tip system Multiple pole tip system Pol e tip stroke length 18 mm 18 mm 18 mm Magnetic tool Tool C (Fig ure 4 8( C )) Tool D ( Fig ure 4 8 ( D )) Iron particles (150 300 m dia.) : 80 wt% Aluminum oxide (WA) magnetic abrasive (<80 m mean dia.) : 20 wt% 12 mg 6.4 mg 2 6.4 mg 2 3 mg 1.6 mg 2 1.6 mg 2
82 S ection 4.3.3 will discuss the characteristics of a 72 mm long tube finished using Tool C and Tool D in the multiple pole tip syste m. T h e finishing experiments using the single pole tip system s will also be discussed for comparison In the case of single pole tip system, to finish the same 72 mm length requires a two step process with 18 mm pole tip feed stroke Figure 4 12 External view of multiple pole tip system (18 18 18 pole tip set) (Photograph courtesy of Junmo Kang) Figure 4 13 s hows the relationship between surface roughness (the average of ten measurements) and distance X in the cases of the multiple pole tip system using Tool C, multiple p ole tip system using Tool D, and single pole tip system. No magnetic tool was used in the single pole tip system (Table 4 2 ). The roughnes s es at X = 0 and 74 mm are measurements of unfinished surfaces. In the MAF process, the finishing operation is performed by masses of mixed type magnetic abrasive aligned with the lines of magnetic force. Due to the unstable mixed type magnetic abrasive motion, the finishing capability generally diminishes at the borders corresponding to the pole tip
83 edges. As a result, the areas around X = 0 36 and 7 2 mm were less finished, which is more clearly shown by the results from the single pole tip system. Figure 4 13 Changes in surface roughness with distance X The material removal realized by the multiple pole tip system using Tool C was 7.85 mg and it was about one third of that per finishing step of the single pole tip system: 12.98 mg (by the first step) and 11.05 mg (second step). T he number of cutting edges active at each finishing area in the Tool C case is estimated to be about half the number in the single pole tip system. The single pole tip system does not use a tool, and the mixed type magnetic abrasive forms chains by magnetic force conforming to the shape of the tube. In turn, the mixed type magnetic abrasive remov es the material not only from the peaks but also the slopes of the surface asperities over the finishing
84 area, as long as the mixed type magnetic abrasive enters the valley of the surface asperities. In contrast, the multiple pole tip system uses a tool ( a solid rod ) and the mixed type magnetic abrasive sandwiched between the tube surface and tool at the regions corresponding to the pole tip edges plays a major role in removing the material. As mentioned above, in the case using Tool C, s ome mixed type mag netic abrasive adheres to the surface of the tube because the friction between the mixed type magnetic abrasive and tube surface exceeds the magnetic force acting on the magnetic abrasive. The region corresponding to the chuck end edge of the pole tip must have the least amount of mixed type magnetic abrasive regardless of the back and forth motion of the pole tip. This lack of abrasive also discouraged the finishing operation. Accordingly, less mixed type magnetic abrasive is involved in the cutting perfor mance in the multiple pole tip system using Tool C than that in the single pole tip system case. The difference in the material removal rate s and surface improvement is attribute d to these differen t mechanisms T he multiple pole tip system using Tool D smo othly finished the surface in an area deep inside the tube corresponding to the pole tip located close st to the chuck. A roughness was achieved similar to that in the single pole tip system case. This resulted from the relative motion of the sufficiently distributed mixed type magnetic abrasive guided by the tool against the tube surface. A fter finishing with Tool D the material removal of 14.2 mg is approximately twice that obtained in the Tool C case, but the material removal per length 0.20 mg/mm is ca lculated about half of the single pole tip system (0.36 mg/mm). The use of a solid tool encourages the amount of mixed type magnetic abrasive participating in the finishing performance and pushes it against the
85 tube surface more strongly than Tool C. Howev er, since the material is pre dominantly removed from the peaks of the surface asperities, the material removal rate must be lower than the case with the single pole tip. H igh values of surface roughness, similar to the unfinished surface are measured at t he center of the finished area. They should be eliminated by enhancing the mixed type magnetic abrasive to contact with the tube surface which can be achieved by extending the pole tip feed Although the finishing result is not shown in this paper, the ro ughness peak exhibited around X =36 mm in the case of the single pole tip system was eliminated by lengthening the pole tip feed to 24 mm in the first phase and 18 mm in the second phase. The 6 mm overlap of the finished area between the first and second ph ases facilitated the material removal from the center area and improved the uniformity of the finished surface. A nalogous ly extension of the pole tip feed length should diminish the unevenness of the roughness in the finished area in the multiple pole tip system. Section 4.3.4 will find a method to determine sufficient pole tip feed length by study ing th e relationship between the pole tip feed, the overlap of finishing area, and the finished surface roughness. 4.3.4 Pole tip Feed Length and Surface Uniform ity The relationship between the pole tip feed length and the finishing characteristics, especial ly surface roughness, are studied here. The finishing experiment starts with the edge of pole tip set ( at X =0 ) as shown in Figure 4 15. The pole tip set move s back and forth according to the feed length. The pole tip takes 61 s for one 18 mm stroke. In this case, the positions X =18 and 54 mm are covered for the longest duration while the positions X =0, 36, and 72 mm experience almost no coverage To cover the center area
86 ( X = 36 mm), the stroke length is increased and the effects of stroke length on finished area is calculated and the surface is studied. Figure 4 14 Changes in pole tip coverage times and surface roughness with distance X
87 Figure 4 14 shows the relationship between pole tip coverage time, roughness of finished surface, and distance X In this study, every 2 mm movement of the pole tip ( corresponding to 3.21 s) is plott ed as the pole tip coverage time in the course of one feed. Figure 4 1 4 ( A ) shows the case with a pole tip feed of 18 mm. The positions with higher pole tip coverage times exhibit lower surface roughness. The increase of the pole tip coverage time increases the contacts between the mixed type magnetic abrasive and the tube surface and thus facilitates material removal. At X =18 and 54 mm, the surface roughness is lower than 0.2 m Rz due to the longest pole tip coverage time 173.4 min under the condition s. In contrast, the positions X =0, 36, and 72 mm exhibit higher roughness due to the lack of contact between the mixed type magnetic abrasive and tube surface. The areas between X =4 and 32 mm and between X =40 and 68 mm have pole tip coverage time s of more than 38.5 min and surface roughness improvement from ~3 m Rz to ~1 m Rz In other words, to achieve a roughness less than 1 m Rz over the entire finished area, the pole tip feed length must be set so as to result in a pole tip coverage time at X =36 mm of at least 38.5 min. T o prove this concept, a pole tip feed of 22 mm was chos en for the experiments. The area from X =36 to 40 mm is the area covered by both pole tips, and the pole tip coverage time there is calculated to be 48.6 min Figure 4 1 4 ( B ) shows the experimental finishing results and the pole tip coverage time plotted aga inst the distance X The roughness peak at X =38 mm is decreased because of the increased pole tip coverage. The roughness was improved to less than 0.8 m Rz If the desired surface is less than 0.5 m Rz a pole tip coverage time greater than 60 min is su ggested by Figure 4 1 4 ( A ). To satisfy this condition, a pole tip
88 stoke of 24 mm is proposed. The 24 mm pole tip feed provides the superposed area from X =36 to 42 mm with a coverage time of 68.3 min As seen in Figure 4 1 4 ( C ), the roughness peak at the cent er is no longer observed and the surface roughness is around 0.2 m Rz Figure 4 15 Three dimensional surface shapes measured by optical profiler Figure 4 15 shows the three dimensional shape s ( measured by an optical profiler ) of the unfinished inner surface and the center of the inner surface s finished with pole tip feeds of 18, 22, and 24 mm. The surface finished with the 18 mm pole tip feed ( Figure 4 1 5 ( B ) ) is similar to t he unfinished surface ( Figure 4 1 5 ( A ) ) Figure 4 1 5 ( C ) with the 22 mm pole ti p feed shows some asperities that remained from the initial surface condition. In the surface with the 24 mm pole tip feed ( Figure 4 1 5 ( D ) ) the
89 asperities from the initial condition are barely observed, and the finished surface consists of cutting marks g enerated by mixed type magnetic abrasive. The pole tip feed conditions of 22 mm (total finished length: 76 mm) and 24 mm (total finished length: 78 mm) had material removal of 16.83 mg and 18.7 mg, and the material removal per length are calculated to be 0 .22 mg/mm and 0.24 mg/mm respectively. Regardless of the extended pole tip feed, the material removal rate of the multiple pole tip system was less than that of the single pole tip system ( 0.36 mg/mm ). This condition resulted from the differences in the finishing mechanisms with and without the use of the magnetic tool. Moreover, i t should be noted that the single pole tip system took twice the time (6 hr) compared to the multiple pole tip system to finish the surface of a 72 mm long tube.
90 CHAPTER 5 D EVELOPMENT OF HIGH SPEED FINISHING MACH INE FOR INTERNAL FIN SHING OF CAPILLARY TUBES 5.1 Design and Construction of High speed Finishing Machine In the internal magnetic abrasive finishing of capillary tubes, the tube rotational speed is a critical factor f or efficient finishing. However, as the tube rotational speed increases, stronger centrifugal force affects the magnetic abrasives and lubricant at the finishing area. To clarify the relationship between tube rotational speed and the finishing characterist ics, new high speed finishing equipment which rotates the tube up to 30000 min 1 has been designed and developed. Previously, the high speed finishing machine with 30000 min 1 successfully finished the inner surface of capillary tube with 0.4 mm ID . However, the multiple pole tip finishing method has not been applied to a high speed finishing process. To clarify the finishing characteristics, a new high speed machine is proposed and constructed. 5.1.1 Proposal of New Finishing Machine The specificat ion s of the finishing machine used in C hapter s 3 and 4 are shown in Table 5 1( A ), and the external view of the machine is shown in Figure 5 1 (finishing unit is shown enlarged in F igure 3 4). The maximum main spindle rotation is limited to 3000 min 1 and t he feed (0 10 mm/s) and vibration (0 5 mm, 0 3 Hz) motion of the pole set are controlled by computer T his rotational speed significantly limits the finishing speed and eventually affects finishing efficiency. This machine can finish tube s which have 0 20 mm OD and are up to 250 mm long In C hapter 4, the multiple pole tip system has been applied successfully to this machine for capillary internal finishing. Even though this system shows a relatively reduced processing time, it still requires long processin g time (3 hrs for 72 mm finishing) [3 5 38].
91 Figure 5 1 External view of the low speed finishing machine (up to 3000 min 1 ) (Photograph courtesy of Junmo Kang) To enhance the finishing efficiency, a high speed internal finishing machine with spindl e speed up to 30000 min 1 was developed in Japan in 2004 T his machine was capable of finishing an 8 mm length of an internal capillary tube (0.4 mm ID tube) from several micrometer s to less than one micrometer in 10 min. Thus, this machine became the mode l for the newly designe d high speed finishing machine T he specifications of the previously developed high speed machine are shown in Table 5 1( B ). Further improvements of this machine design were considered, and the following two issues were mainly consid ered to improve the process ing performance : ( 1) the main spindle rotational speed and ( 2) the control system for positioning the magnetic pole tip s The main spindle determines the finishing speed and is attached to the workpiece chucking system. In the ca se of the machine in Figure 5 2 the main spindle rotation 10 cm Motor Plastic cover Rigid support Pole set 3 jaw chuck
92 was controlled by a pneumatic gearbox motor, and the rotational speed s ( 4070, 7653, 30000 min 1 ) w ere adjusted by the gear combination Th is machine has only three gear combination s which limits t he variation of the workpiece rotational speed. To overcome th is limitation a main spindle with variable rotationa l speed was desired. Table 5 1. Specifications of finishing machines; ( A ) Existing high speed machine at UF, ( B ) Existing high speed machine, ( C ) Newly proposed high speed machine In capillary tube finishing the magnetic field is a key parameter to control the magnetic force, which generates the finishing force in the process. In the machine in Figure 5 2, the magnetic field is generat ed by permanent magnets, and the position al relationship between the finishing area and pole tips must be critically controlled to within an order of 10 m in order to obtain the desired magnetic field intensity. This is more important and difficult as the tube diameter becomes smaller. The clearance between the pole tip and pole position was practically determined by using a spacer of known thickness ( e.g. polytetrafluoroethylene (PTFE) tape ) Because t he roundness ( A ) Existing machine at UF (Low speed machine) ( B ) Existing high speed machine ( C ) Proposed machine (High speed machine) Maximum workpiece size 20 mm outer diameter 25 0 mm length 4 mm outer diameter 15 0 mm length 3 mm outer diameter 20 0 mm length Main spindle rotation 0 3000 min 1 Programmably controlled 4070, 7653, 30000 min 1 500, 5000 3 0 000 min 1 Programmably controlled Maximum pole feed rate 10 mm/s 80 0 mm/s 60 0 mm/s Pole vibration Maximum amplitude : 5 mm Maximum frequency: 3 Hz Pole tip positioning Relative to w orkpiece surface Relative to w orkpiece surface Absolute position control Machine dimensions H2 7 0D 46 0W 51 0 mm H2 0 0D 2 50W 45 0 mm H2 7 0D 35 0W 50 0 mm Weight 1 5 kg 1 0 kg 1 3 kg Power supply 1 10 VAC, 50/60 Hz 110 VAC, 50/60 Hz 110 VAC, 50/60 Hz
93 and cylindricity of the workpiece were not critically controlled in the original manufacturing process find ing the center of the workpiece was difficult. Moreover, misalignment of the pole tips would result in the instability of the abrasive motion and the nonuniformity of the finished s urface. In the existing low speed machine (Figure 5 1), it is difficult to adjust the pole tip distance because the positioning work has to be done by manually and once the pole tip is stuck together every set up has to be done again. To overcome this iss ue, the new machine was desired to have a precision pole tip positioning system. Figure 5 2 External view of existing high speed finishing machine  Based on these concepts, the design specifications were determined as show n in Table 5 1( C ). The electric motor to control the main spindle was s elected because i t has high torque to accommodate the larger and heavier chucks required for larger wor k pieces, and tasks ranging from simple rinsing to heavy duty cleaning ( including d eburring ) can be performed on various diameter workpiece s The rotational speed s of the motor are in the range of 5000 30000 min 1 and 500 min 1 To simplify the machine, Nd Fe B permanent magnet (18 12 10 mm) Workpiece Linear slide Yoke Motor Pole tip 390 mm 180 mm CCD camera
94 t he pole vibration system was not considered To allow the pole tips to be positioned to within 10 m with respect to the tube in radial direction t he permanent magnet and pole tips are mounted on a precision single axis micrometer stage. Figure 5 3 Schematic of the proposed high speed finishing machine Figure 5 3 sh ows the schematic of the proposed high speed machine It was designed using Solidworks to avoid interference between the parts. The main spindle rotation and the pole tip motion are controlled through software, and t he pole tip geometry can be modified depending on the wor kpiece size. 5.1.2 Description of N e w High speed Finishing Machine Based on the prototype in Figure 5 3, the new high speed finishing machine is developed. Figure 5 4( A ) shows overview of the machine developed. The machine consist s of a finishi ng unit (Fig ure 5 4( B )) and a separate control unit (Figure 5 4( C )) which are located beneath transparent cover s for safety reason s The control unit is connected to the finish ing unit to control the main spindle and pole motion s The Motor Linear slide Jig Workpiece 500 mm 350 mm 270 mm Single axis stage Pole tip
95 necessary motion is programmed i n advance using the computer which is connected to linear guide controller, and it is transferred through the control unit to the finish ing unit. The air l ine kit keeps the main spindle clean during operation by providing a continuous flow of air ( 0. 15 0. 25 MPa), which prevents the accumulation of dust in the small clearance between the main spindle and spindle holder. Moreover, the compressed air cools the spindle assembly and machine during operation. Figure 5 4 Photographs of developed machine overvi ew, finishing unit, and control unit (Photograph courtesy of Junmo Kang)
96 Figure 5 5 Photograph of finishing unit and pole tip geometry (Photograph courtesy of Junmo Kang) The finishing machine is just like the schematic shown in Figure 5 3. The neodymi um iron boron permanent magnets ( Grade N42, 0.5 0.5 0.5 in (12.7 12.7 12.7 mm)) are used to generate the magnetic field [ 67 ]. T he magnetic flux density of 0.42 T was measured with a H all sensor at the center of the magnet surface. The pole tip geometry sho wn in Figure 5 5 was designed to generate a magnetic field strong enough to generate relative motion between the magnetic abrasive and the workpiece inner surface. Practically, the pole tip geometry is determined by the workp ie ce size. For capillary finish ing, each pole tip was attached to three magnets, and the magnetic flux density of 0.14 0.15 T was measured at the pole tip end. Since the detecting area of the H all sensor (1.0 mm diameter) exceed ed the pole tip end area
97 (0.5 4 mm), the H all sensor showed less magnetic flux density. The magnetic flux density at the pole tip end is estimated theoretically to be over 1 T . The magnet and pole tip s ubassemblies can be manually moved in the tube radial direction and they are manipulated in the tube axial direction by a linear slide T he motor controller and timer relay s shown in Figure 5 4( C ) control the reciprocation of the pole tips : the pole feed rate and the start and end position s. 5.2 Finishing Characteristics of High speed Machine Finishing experime nts of stainless steel capillary tubes (0.64 mm OD, 0.48 mm ID) were executed with the newly developed machine ( F igure 5 5, s ingle pole tip system) at a spindle speed of 30000 min 1 to confirm the machine s finishing performance Table 5 2 shows the experi mental conditions of existing low speed and developed high speed finishing machines In the case of the new machine, the finishing experiments with various time s have been executed to produce the finely finished surface (~0.1 m Rz ). Since the experiments did not require significant material removal from the workpiece, diamond abrasive was not considered A mixture of magnetic abrasive and ferrous particles was supplied as a cutting tool. F inishing experiments with the existing low speed machine ( F igure 5 2 ) were also conducted to compare the finishing performance of the two machines. Theoretically, the finishing time for high speed machine can be roughly estimated using the previous experiment conducted using the existing machine. The number of stroke s is 180 at 2500 min 1 However, the rotational speed has increased with developed machine twelve times so the number of stroke s has been considered twelve times less than the previous result.
98 Figure 5 6 shows the microscopy of the tube interior of the unfinish ed and finished surfaces. In both cases, the rough ly drawn surfaces were modified to ones covered with unidirectional scratches generated by the magnetic abrasive during finishing The vertical direction is the circumferential direction of the tube. It was confirmed that the developed high speed machine was capable of finishing the inner surface of the capillary tube at 30000 min 1 in less than one fifth the time needed for low speed finishing. Table 5 2 Experimental condition s for finishing performance be tween existing low speed machine and newly developed high speed finishing machine C lose observation of the finished surfaces show that the cutting marks in the 2500 min 1 tube rotation case (Figure 5 6(A ii )) are more uniform than the ones in the 30000 min 1 case (Figure 5 6(B ii )) The high speed rotation is twelve times faster than the low speed case D ue to the higher centrifugal force i n the high speed case, t he lubricant mixed with the mixed type magnetic abrasive spun off easily The high er the tube rotational speed, the easier the ejection of the lubricant from the finishing area. The lubricant plays an important role by ( 1) encouraging the mass of mixed type magnetic abrasive to conform to the inner surface of the tube at the beginning o f the process and Machine Existing machine (Low speed machine) Developed machine (High speed machine) Workpiece 304 stainless steel tube : 0.64 0.48 60 mm Workpiece revolution 25 00 min 1 30 000 min 1 Mixed type magnetic abrasive Iron particles ( 50/+100 mesh, 150 300 m dia.): 0.4 mg + Aluminum oxide (WA) particles (80 m mean dia.) in magnetic abrasive (<10 m): 0.1 mg Workpiece Pole clearance 0.11 mm Pole feed Feed length : 4 mm; Feed ra te : 0.5 mm/s Lubricant Soluble type barrel finishing compound Processing time 54 min 10 min
99 ( 2) maintaining smooth relative motion against the tube surface. T he lack of the lubrication deteriorates the smooth relative motion between the magnetic abrasive and the tube inner surface, resulting in irregular cutting marks I n additi on lubricant can cool down the temperature during the cutting process, so the lack of lubrication may cause the adherence of particles onto the surface due to the friction. Figure 5 6 Microscopy of tube interior s before and after finishing at the rotat ional speeds of 2500 min 1 and 30000 min 1 Figure 5 7 shows representative three dimensional shapes of the inner surface of tubes. The finished surfaces are typical in the MAF process, and the irregular micro asperities are more visible in the case of high speed finishing than with low speed finishing As mentioned above, this must be due to the mixed type magnetic abrasive behavior as well as the lack of lu bricant at the finishing area.
100 Figure 5 8 shows measured surface roughness of unfinished and finished surfaces in both conditions (Table 5 2) The initial surface conditions of the stainless steel tubes depend on the tube manufacturing processes, and the roughness of drawn 304 stainless steel tubes is typically around 7 m Rz After 54 min in the case of low speed conditions and 10 min in the case of high speed conditions respectively the tube surfaces were smoothly finished to 0. 1 0. 3 m Rz It is seen that the surface finished with high speed conditions shows slightly higher roughness than the one wit h low speed condi tions. Figure 5 7 Three dimensional surfaces of tube interiors before and after finishing at the rotational speeds of 2500 min 1 and 30000 min 1 measured by optical profilometer
101 Consequently, the finishing experiments demonstrated the a pplicability of the developed machine for high speed internal finishing of capillary tubes. The results shown demonstrated that the developed high speed machine exhibits improve d finishing efficiency over 5 times compared to the existing low speed machine. As mentioned above the previously developed high speed machine in Japan required 8 min to improve the surface roughness of 304 stainless steel tubes (0.5 mm OD, 0.4 mm ID) from 2 m Rz to 0.15 m Rz . Although the tube size is slightly different, it can be confirmed that the newly developed machine can perform the internal finishing with superior speed. This is a result of the precision pole tip position control, which achieves precision control of magnetic field at the finishing area. Figure 5 8 S urface roughness of unfinished and finished surfaces in rotation al speed 2500 and 30000 min 1
102 CHAPTER 6 HIGH SPEED INTERNAL FINIS HING OF CAPILLARY TU BES BY MULTIPLE POLE TIP SYSTEM 6.1 Internal Finishing of Capillary Tubes by High speed Multiple Pole tip System In order to improve the finishing efficiency, a multiple pole tip system has been proposed and applied to the internal finishing process. The development of a multiple pole tip system using a partially heat treated magnetic tool allows the finishing of multiple regions simultaneously in capillary tubes. Moreover, the high speed finishing machine has also been developed to improve finishing efficiency and demonstrate its finishing performance. Although an increase in finishing efficiency in multiple p ole tip system has been observed, the finishing characteristics of general straight capillary tubes in the n ewly developed high speed machine has no t been studied yet. Therefore, high speed internal finishing of capillary tubes using a multiple pole tip sy stem has been developed and the respective finishing experiments ha ve been executed. Firstly, finishing equipment with double pole tip sets is developed, which enables a tube to r otate up to 30000 min 1 Secondly, t he effects of tube revolution on abrasive motion are investigated t hrough the tube finishing experiments. The behaviors of the tool and magnetic abrasive as a function of tube rotational speed are studied. In addition the finishing characteristics and mechanism are studied using the straight cap illary tubes Finally, the flexible capillaries are applied to the high speed multiple pole tip finishing method and studied its finishing characteristics and mechanism Figure 6 1 shows a schematic for a method using double pole tip sets, which generates magnetic fields in two finishing area s and a photograph of the equipment developed to realize the method. The finishing area is doubled as magnetic abrasive is introduced and pushes against two regions of the tube surface. As the pole tip set s
103 move along the tube axis, the finished area is extended. The number of pole tip sets can be increased if needed. For a constant pole tip width, the finishing area will be a function of the total number of pole tip sets. The double d pole tip sets require the introduct ion of a magnetic tool with the mixed type magnetic abrasive. The tool guide s the magnetic abrasive deep into the tube and increase s the magnetic force acting on the magnetic abrasive as mentioned in Chapter 4 Figure 6 1 Schematic of the processing pri nciple for internal finishing by multiple pole tip system (Photograph courtesy of Junmo Kang) The workpiece (straight capillary tube ) is chucked to a motor ( speed range : 500 min 1 5000 30000 min 1 ). Two pole tip sets consisting of s ix neodymium permanent magnets (12.7 12.7 1 2.7 mm ; r esidual f lux density 1 26 1 .2 9 T ; c oercive f orce > 875 AT/m) are mounted in right angle configuration on a single axis micrometer stage, and their position is adjustable in the tube radial direction. To avoid collision between t he rotating tube and pole tip s the pole tip surface s are covered by 0.3 mm thick
10 4 polytetrafluoroethylene (PTFE) tape. The pole tip set s are mounted on a linear slide so that they can be fed in the tube axial direction T h e feed length and speed are adjust able to maximums of 150 mm and 600 mm/s respectively Figure 6 2 Changes in magnetic flux density in multiple pole tip sets at Y =0 mm Figure 6 2 shows changes in magnetic flux density B y measured by a Hall sensor (sensing area: 1.0 mm), with distanc e X for double pole tip sets The magnetic flux density and gradient increases from the center toward the edges of pole tip. A particle in the magnetic field is attracted to the pole tip edges where the magnetic force is higher Table 6 1 shows the experim ental conditions. Austenitic stainless steel tubes ( 304 stainless steel, 1.27 1.06 100 mm; 2 3 m Rz initial surface roughness) were prepared as workpieces for this experiment A 304 stainless steel tool with three heat
105 treated regions was used as a magn etic tool (Fig ure 6 3). The heat treated regions became non magneti c as a result of the treatment while the f our un treated sections remain ed magnetic The mixed type magnetic abrasive separate s as it is attracted to the ends of the four magnetic sections. The magnetic regions correspond to the two sets of magnetic pole tip s The pole tip feed length was set to 12.7 mm, and the feed rate was set to 0.59 mm/s Table 6 1. Experimental c onditions for the high speed internal finishing of straight capillary tube s For the experiments 5 mg of mixed type magnetic abrasive (80 wt% iron particle s and 20 wt% magnetic abrasive) was introduced with the magnetic tool The mixed type magnetic abras ive and magnetic tool filled 21.4 vol % and 23.1 vol % inside the tube, respectively The tube rotational speeds were varied between 5000, 10000, 20000 and 30000 min 1 T o encourage the mixed type magnetic abrasive to uniformly cover the tube surface prior to high speed finishing, the tube was rotated at 500 min 1 with a single pole tip stroke before finishing. Workpiece 304 Stainless steel tube ( 1.27 1 06 100 mm) Workpiece revolution 500, 5000, 10000, 200 00 and 30000 min 1 Mixed type magnetic abrasive Iron particles ( 50/+100 mesh, 150 300 m dia.): 4 mg + Aluminum oxide (WA) particles (80 m mean dia.) in magnetic abrasive (<10 m): 1 mg Magnetic tool Fig ure 6 3 Pole tip feed 0.59 mm/s Pole tip feed l ength 12.7 mm and 16 mm Workpiece pole tip clearance 0.3 mm (P olytetrafluoroethylene (PTFE) tape thickness) Lubricant Soluble type barrel finishing compound (pH : 9.5, Viscosity: 755 mPas at 30C) Processing time 10 and 20 min
106 Figure 6 3 Tool geometry and magnetized tool with iron particles (Photograph courtesy of Junmo Kang) 6.2 Effects of Lubrication on Finishing Char acteristics In high speed finishing, centrifugal force tends to cause displacement of the lubricant from the finishing area, which in turn causes the mixed type magnetic abrasive to adhere to the tube surface as a result of friction and heat Adhered mat erial noticeably cove re d the surface in Fig ure 6 4 ( B ), which shows the surface finished continuously for 10 min at a tube revolution rate of 30000 min 1 No adhered material is observed in Fig ure 6 4 ( C ), which shows the surface finished for 10 min with lub ricant added after 5 min. During the finishing process, the presence of lubricant is crucial to encourage the smooth relative motion between the mixed type magnetic abrasive and the tube surface that facilitate s finishing performance of the abrasive Accor dingly, i n the case s of 20000 and 30000 min 1 tube revolution, the finishing experiments were interrupted to inject fresh lubricant every 5 min.
107 Figure 6 4 Microscopy of the initial and finished surface for 10 min at 30000 min 1 : Initial surface finish ed for 10 min continuously and finished for 5 min and ano ther 5 min with fresh lubricant F or the cases of 5000 and 10000 min 1 tube revolution, the finishing experiments were performed continuously for 10 min without additional lubricant injection Each experiment was repeated at least three times to ensure the repeatability of the results. Before and after the finishing experiments, the tube was rinsed with ethanol in an ultrasonic cleaner for 1 hr to measure the material removal rate
108 6.3 Effects of Tub e Rotational Speed on Finishing Characteristics Figure 6 5 and 6 6 show intensity maps and oblique plot s measured by an optical profiler at X =13 mm of the unfinished surface and surface s finished for 10 and 20 min, respectively Figure 6 8 shows changes in material removal with tube revolution and finishing time The surface finished for 10 min at a tube revolution of 5000 min 1 (Figure 6 5( B )) has a roughness of 0.15 m Rz but multiple irregular asperities from the initial surface remained. However, an ex tension of the finishing time for another 10 min allowed the abrasive to remove the irregular asperities. The material removal after 20 min was more than twice the removal after 10 min. Although the magnetic abrasive was not exchanged during the interrupti ons to inject lubricant the reconfiguration of the magnetic abrasive during these breaks encouraged the re location of sharp abrasive cutting edges. The sharp cutting edges and newly added lubricant seemed to refresh the finishing performance after each in terruption At a tube revolution rate of 10 000 min 1 the surface was smoothly finished (0.1 m Rz ) after 10 min, and the roughness value remained constant after 10 min of extra finishing time despite the additional material removal (Fig ure 6 8) C ompared to 10 min at a tube revolution rate of 10000 min 1 t he material removal is drastically increased after finishing for 20 min. Analogous to the case of finishing for 20 min at 5000 min 1 the pause s to add lubricant after 10 min aided the finishing performa nce. It was confirmed that the increase in the tube revolution (i.e., cutting speed ) improves the material removal rate and finishing efficiency. However, due to the high centrifugal force, th e high speed tube rotation creates more opportunities for the mi xed type magnetic abrasive and magnetic tool to lapse into unstable conditions The lack of a uniform magnetic abrasive distribution under an unstable rotating magnetic tool may
109 lead to the deep irregular scratches on the surface. This trend was observed in the case of 20000 min 1 (Figure 6 5( D )), the finished surface has deep scratches and surface distortions (Figure 6 5( E )). E xtending the finishing time slightly increased the material removal due to the longer duration contact of the magnetic abrasive cu tting edges against the tube surface and removed the relatively short wavelength surface asperities (Figure 6 6( C )) ; however, t he deep scratches produced by the irregular motion of the magnetic tool and mixed type magnetic abrasive remained on the surface. In Figure 6 7, irregular surface asperities were observed at the rotational speeds between 2 0000 and 3 0000 min 1 Especially, in the case of 30000 min 1 after 10 min finishing, the surface roughness is almost 10 times higher (1.12 m Rz 5.52 mg ) than t he cases of 5000 ( 0.15 m Rz 1.5 mg ) and 10000 min 1 (0.10 m Rz 2.02 mg ) ; despite a few times material removal (Figure 6 8) The finished surface with irregular surface asperities, in spite of higher material removal, must be related with instability o f magnetic tool and abrasive behavior due to higher centrifugal and friction force. Moreover, after 100000 times of tube rotation at 5000 and 10000 min 1 surface finished at 10000 min 1 produced the smoother surface (0.10 m Rz ) with less material removal (2.02 mg) compared with slower rotation case (0.12 m Rz 4.52 mg) On the other hand, after 200000 times of tube rotation in each rotational speed, the surfaces finished at 5000 and 10000 min 1 have similar material removal (8.2 and 7.69 mg respectively) Conversely the irregular asperities were remained the surface finished at 20000 min 1 with less material removal (0.36 m Rz 5.38 mg) This indicates t he material removal per tube rotation is decreased with increasing tube rotational speed.
110 Figure 6 5 Intensity maps and oblique plots of the initial surface and the surface finished for 10 min at 5000 min 1 10000 min 1 20000 min 1 30000 min 1
111 Figure 6 6 Intensity maps and oblique plots of the surface finished for 2 0 min at 5000 min 1 10000 min 1 20000 min 1 Use of the pr eviously developed multiple pole tip finishing system produced a uniformly finished surface (from 2 3 m Rz to ~0.2 m Rz ) 72 mm long four times the pole tip width in 180 min (at 2500 min 1 ) The high speed finishing system prop osed in this research can produce a finished surface 50.8 mm long ( four times the pole tip width ) in 10 min with a roughness of about 0.1 m Rz Thus the newly developed high speed finishing system is twelve times more efficient than its predecessor Inter nal machining of flexible capillar ies with laser machined slits w ere performed using the high speed finishing machine. The processing characteristics will be presented in Section 6.4
112 Figure 6 7 Changes in surface roughness with tube revolution observe d in multiple pole tip system Figure 6 8 Changes in material removal with tube revolution in multiple pole tip system 6.4 Flexible Capillary Tube Finishing 6.4.1 Finishing Characteristics Figure 6 9 shows a schematic of processing principle for interna l finishing of flexible capillary tube by multiple pole tip finishing system. The partially heat treated austenitic stainless steel tool and mixed type magnetic abrasives are introduced into the tube U nder effects of magnetic field both tool and abrasive s are attracted into the inner
113 surface of the tube. W hen the tube rotates with reciprocating motion of the pole tips, the surface and edges inside the tube can be finished gradually. Figure 6 9 Schematics of the processing principle for the internal fin is hing of flexible capillary tube s by a multiple pole tip system Two differently heat treated tools ( T ool C and D, Figure 6 10( A ) and ( B ) respectively) are prepared for the finishing experiments. A photograph of the equipment is shown in Figure 6 11 for t he internal finishing using double pole tip sets The flexible capillary tubes with laser machined slits were prepared. In order to prevent the irregular tube rotation the free end of the flexible capillary tube is wrapped with PTFE tape and held by the j ig. Figure 6 10 Schematics of the partially heat treated metastable austenitic stainless steel tool s: T oo l C and T ool D
114 Figure 6 11 External view of high speed multiple pole tip finishing system for flexible capillary tube finishing (Photograph co urtesy of Junmo Kang) Table 6 2 Experimental conditions for the high speed internal finishing of flexible capillary tubes : Tube B Workpiece Flexible stainless steel tube ( 1.36 1.02 100 mm) Workpiece revolution 10000 min 1 Pole reciprocating motion Sp eed: 0.5 mm/s, Stroke length: 16 mm Lubricant Soluble type barrel finishing compound (pH: 9.5, Viscosity: 755 mPas at 30C): Fresh lubricant addition every 6 .5 min (6 stroke times) Mixed type magnetic abrasive Iron particles ( 50/+100 mesh, 150 300 m d ia.): 80 wt% + Aluminum oxide (WA) particles (80 m mean dia.) in magnetic abrasive (<10 m): 20 wt% ; 5 mg Diamond abrasive 50 70 m diamond paste Magnetic tool Fig ure 6 10 To increase the pole tip coverage time at the center of finishing area, the s troke length wa s set to 16 mm The overlap covers the less finished surface due to friction between abrasive and inner surface. In the case of internal deburring process with
115 flexible capillaries the lubricant easily escapes through the slits during tube rotation, and higher material removal rate is require d compared with surface finishing of straight capillary tube. For encouragement of the material removal, l ubricant mixed with diamond abrasive s (50 70 m) wa s added every 6 times of pole stroke (Table 6 2) A fter the micro laser machining process, the re solidified material adhered on the edge and surface inside the tube The initial surface has burrs on the edges which have a height up to 90 m ( Figure 6 1 2 ( A ) ). This is slightly bigger than the size of t he magnetic abrasive. T hese burrs obstruct the magnetic abrasive to follow the pole reciprocating motion. In order to make the abrasives go over the burrs, the heat treated magnetic tool is introduced to increase the magnetic force acting on the abrasives. Figure s 6 1 2 ( B ) and 6 1 2 ( C ) represent the surfaces finished by T ool C and T ool D for 26 min respectively. The burrs on the edges and re solidified material on the surface we re completely removed from the surface finished by T ool C. Conversely the burrs on the edge and the re solidified material re m ain ed on the surface finished by T ool D Figure 6 1 3 shows the spectrum plots of as received surface and surfaces finished by T ool C and T ool D The surface roughness indicated in the each plot is measured in t he average of 10 measurements with 100 m length surface profiles. As shown in Figure 6 1 3 ( A ), the i nitial surface roughness between slits is measured in 2.5 3.5 m Ra The surface finished by T ool C (0.3 0.4 m Ra ) is rougher than the case finished by too l D (0.2 0.3 m Ra ). This trend was observed in previous result in Section 3.3.3. This demonstrates that the developed high speed multiple pole tip finishing system delivers the magnetic abrasive inside the flexible capillary tubes.
116 Figure 6 1 2 Microsco py of tube in t eriors: As received condition and s urface finished for 26 min by T ool C and T ool D
117 Figure 6 1 3 Spectrum plots of the tube in teriors: As received condition surface finished by T o ol C and T ool D
118 According to the pr evious experimental result s (2.3 mg 8 mm finish ed length ) describe d in Section 3.3.3 0. 2 9 mg/mm of material removal was expected for 54.1 mm finished length In the high speed finishing system, t he case of T ool C, it achieved 0.24 mg/mm, which is close t o the desired removal rate. However, the case of T ool D showed only 0.15 mg/mm This impl ies that T ool C is more appropriate for high material removal in a short er time period while the T ool D is slower in material removal but producing smoother surface. Figure 6 1 4 Microscopy of tube ex teriors: As received condition and surface finished for 26 min Figure s 6 1 4 show s the external view of flexible capillary tubes before and after finishing process for 26 min respectively. Even though the gap s between th e slits were smaller ( ~ 12 m) than the abrasive diameter, the abrasive s we re crushed and finished debris could be leaked out from slits under the effect of finishing pressure T he leaked particles come into contact with outer surface during finishing proce ss and cause the material removal on the outer surface along the tube axis The T ool C has achieved internal finishing of flexible capillary tubes in timely manner but failed to internal finishing of straight capillary tubes. On the contrary to this,
119 T oo l D has produced smoothly finished surface in straight capillary tubes but not internal finishing of flexible capillary tubes. T h e only difference between the tools is heat treated regions (non magnetic areas). This causes different tool behaviors, particl e dis tribution, and magnetic for ces acting on magnetic abrasive and these factors affect on the finishing characteristics critically. S ection 6.4.2 discusses the tool behavior and particle distribution. 6.4. 2 Tool B ehaviors and Particle Distribution In or der to observe the behavior s of each magnetic tool with ferrous particles, the tool and particles are introduced into the transparent glass tube ( 6 9 5 0 75 mm ). In Figure 6 15( A ), due to the uniform magnetic field on the pole tip, the particles are dist ributed along the pole tip width and the tool is strongly attracted against the inner surface. O n the other hand, in Figure 6 15( B ), the particles are mostly divided into four sections which are magnetic regions of magnetic tool and are attracted to the se sections while barely distribut ing a round the centr al area of each pole tip. Moreover, most particles are attracted on the surface strong er than the tool and pushed by the t ool Using the smaller glass tube ( 2 75 2 11 9 0 mm) the tool and particle beh aviors were introduced into the tube and observed during the tube r otation between 5000 and 10000 min 1 In the case of T ool C ( Figure 6 16(a) ) the tool and particles were u nstable even though the rotational speed wa s low ( 5000 min 1 ) The tool was period ically vibrated along the tube axis and rotated itself while the pole set remain s in same position The tool rotation might be caused by friction force between the tool and inner surface during tube rotation. The tool is strongly attracted to the inner sur face and the particles are attracted to the surface around the tool through the line of magnetic
120 field. However in high speed tube rotation (10000 min 1 ), the tool lapsed into u nstable condition (e.g. vibration and rotation), the particles circumvent the tool and the m ajority resituate to the area behind the tool During the process, the tool pushes a few of magnetic abrasives strongly and this makes deep scratches on the surface Figure 6 15 Powder distribution with differently heat treated magnetic tools: T ool C and T ool D (Photograph courtesy of Junmo Kang) Figure 6 16 Schematics of the powder distribution with differently heat treated magnetic tools: T ool C and T ool D
121 I n the case of T ool D ( Figure 6 16( B ) ) both tool and particles shows stable a t the rotational speed of 5000 min 1 T he particles are attracted to the inner surface strongly and the tool pushes the particles in a stab le motion neither vibration nor rotation Moreover, even though the tube rotate d at the speed of 10000 min 1 ; both th e tool and particles retain rigidly and smoothly follows the pole tip motion 6.4. 3 Relationship between F inishing F orce and M agnetic T ools T he heat treatment alters the magnetic properties of the 304 stainless steel tool and the magnetic properties of th e tool determine the abrasive behavior and distribution and tool behavior T o clarify the finishing mechanism of flexible capillary tube s using the heat treated magnetic tools in a multiple pole tip finishing system, the measurement of the magnetic force a cting on the magnetic abrasive is indispens a ble Figure 6 17 Schematics of magnetic force measurement system The system for magnetic force measurement using strain gauge s was designed (Figure 6 17) and implemented (Figure 6 1 8 ). Strain gages ( gage fact or: 2.09 1.0%, gage resistance ( at 24 C, 50% RH): 119.8 0.2 ) were attached in a half bridge arrangement to the aluminum plate (14.6 3.1 150 mm ) and t he plate was located on a three axis micrometer stage In order to avoid physical contact between the bottom of p late and the top of pole tips, a clearance of 0. 3 mm wa s select ed T he heat treated tool
122 with particles (mixed type magnetic abrasive) was located in the groove on the plate to prevent dispersion of the particles As shown in Figure 6 1 8 the particles wer e distributed with both T ool C and D. The magnetic force F acting on the tool with and without particles was measured by two strain gages and e ach measurement was conducted three times. Figure 6 1 8 Photograph of magnetic force measurement system for mu ltiple pole tip finishing machine T ool C with particles, and T ool D with particles (Photograph courtesy of Junmo Kang)
123 Figure 6 1 9 Relationship between magnetic force and ferrous tool Figure 6 19 shows that t he magnetic force acting on T ool C without particles is 1.8 times stronger than that of T ool D without particles T he magnetic area of T ool C is 1.8 times long er than the area in T ool D ( Figure 6 10 ). Increasing the magnetic area increase s its volume, resulting in greater magnetic force in T ool C t han T ool D Adding the particles (5 mg per pole tip, total: 10 mg, Figure s 6 18( B ) and ( C )), the magnetic force acting on the tool and particles wa s increased by factor s of 2.7 and 3.2, respectively, when compared with the cases of T ool C or T ool D only. I n the case of T ool D, the particles were attracted by the smaller areas This increased the particles per area influenced the magnetic field intensity and its gradient, and increase d the magnetic force in the case of T ool D more than the case of T ool C F igure 6 20 shows the relationship between finishing force per area (pressure) and magnetic tools. The contact area was determined by the visual observation as shown in Figure 6 16 Th e pressure s of T ool s C and D without particles were similar but Tool D w ith particles showed higher pressure than T ool C with particles This might be attributed to the differences in particle distribution The stronger pressure led to the
124 stable motion of T ool D, deliver ed the particles deep into the straight capillary tubes, and achieve d the surface finishing over the entire finish ed area On the other hand, the low er pressure of T ool C caused instability of the tool behavior and difficulties in delivering the particles deep into the finishing area As a result, T ool C could not finish the inner surface the tube. Figure 6 20 Relationship between finishing pressure and heat treated tools In the case of flexible capillary tube finishing, T ool D showed a lower material removal rate (0. 29 mg/min) when compared to T ool C (0. 50 m g/min) Moreover T ool D partially left burrs on the edges while T ool C remove d burrs and re solidified material completely The total material removal in 26 min was 7.54 mg and 13.03 mg respectively in the cases of T ool s D and C As mentioned above, T oo l D had more particles per magnetic area s o f the tool. T his m ight have increase d the probability of particles clogg ing inside the tube and retarded the relative motion between the particles in flexible capillary tubes for surface finishing and deburring du e to the existence of
125 burr s In contrast, T ool C successful ly remov ed material, including burr s, and finished the surface despite the low pressure. Some other parameters must have influenced the finishing performance in the case of Tool C which needs to b e clarified to understand the finishing mechanism The straight capillary tubes used in the experiments were most likely magnetized during the cold working production process. S ection 6.4.4 will discuss t h e effects magnetic properties of the tubes on the f inishing characteristics and the finishing mechanism of T ool C will be clarified. 6.4.4 Effects of Tube Magneti sm on Finishing Characteristics Generally, internal finishing of nonmagnetic tubes by MAF is performed in magnetic fields with 0.2 1 T. If a ma gnetic tube is placed instead of nonmagnetic tube for internal finishing m agnetic fluxes tend to flow into magnetic tube in a magnetic field due to the higher permeability of magnetic objective when compared to air. T he magnetic abrasive adheres to the in ner surface of the tube and rotates with the tube when the tube rotates : No finishing action is performed H owever, if the magnetic field intensity is increas ed to magnetically saturate the tube (magnetic flux density above 1 T) the magnetic flux leaks fr om the tube. If the magnetic field inside the magnetic tube becomes strong enough to hold the magnetic abrasive at the finishing area while the tube rotates the inner surface of magnetic tube can be finished A p revious report showed th at MAF enabled the internal finishing of 0.4 mm thick carbon steel tube . The straight and flexible capillary tube s used in this study ha ve thi nner wall s ( 0.105 mm and 0.26 mm respectively ).
126 Figure 6 21 2 D m a gnetic field analysis by FEM with non ferrous and ferrou s tubes
127 In order to estimate the magnetic flux density at the finishing area, a two dimensional finite element method (FEM) was applied. T he magnetic propert ies of 1018 carbon steel were used in place of th ose of a magnetized 3 04 stainless steel Figure 6 21 shows the FEM result of magnetic flux density in non magnetic 304 stainless steel tube and magnetic 1018 carbon steel tubes of 0.105 and 0.26 mm tube wall thickness. The lines in each figure indicate the magnetic flux flow In the case of non magnetic tub e with 0.1 05 mm tube thickness (Figure 6 21(A)) the magnetic flux density at the finishing area is calculated to be 1.23 T. I n the case of magnetic tube with 0.1 05 mm wall thickness (straight capillary tube, Figure 6 21(B)) ; the magnetic flux density is l ower: 1.10 T. This means the tube is almost magnetically saturated and some magnetic flux leak in side the tube Assume a magnetic tool and particles are introduced into the magnetic tube. If the magnetic tool and particles have magnetic permeability high enough to generate magnetic force to overcome the friction between the particles and inner tube surface, finishing action is performed If not, some particles adhere to the tube surface and rotate with the tube, resulting in u nstable finishing Accordingly t he condition with T ool D, which produced higher pressure, more successfully demonstrated tube finishing than the condition with T ool C. I n the case of flexible capillary t ubes ( F igure 6 21 ( C ) and (D)), the lower magnetic flux density in the finishing ar ea wa s calculated in the magnetic tube (0.61 T Figure 6 21(D ) ). T he thicker the tube wall, the weak er the magnetic field in side the tube (at the finishing area ). The flexible capillary tube was manufactured by a combination of cold working and laser mach ining processes. T o create multiple slots s o me sections of tube were melted
128 by the laser machining process. Under th e condition above Curie temperature t he structure s previously transformed from austenitic to martensitic by cold working were transformed back to austenitic structures and the tube partially lost its magneti sm The pressure of T ool C with particles was strong enough to remove the nonmagnetic burrs from the slot without clogging the tube and to finish the surfaces between the slots of the fl exible tube In order to confirm the differences in the tube magneti sm between the straight and flexible capillary tubes, the following experiments were performed. Figure 6 22 Relationship between magnetic flux density and both capillary tubes at dista nce X It was hypothesized that the magnetic flux density between the tube and magnet could be higher i f the tube show ed higher magnetic permeability. Figure 6 22 shows the
129 schematic of magnetic flux measurement system and the changes of magnetic flux densi ty with straight and flexible capillary tubes. To prevent bending, the tubes were put into a glass tube ( 2 75 2 11 9 0 mm) and the glass tube was chucked. The clearance between the glass tube and pole tip was set at 2.5 mm ( z axis). The hall sensor wa s lo cated on the pole tip (underneath the glass tube) and the magnetic flux density was measured in each case : without the tube, and with straight and flexible capillary tubes at X =2 and 6 mm corresponded to the areas close to the pole tip edge and center of pole tip respectively. The strong est magnetic flux density was detected with the straight capillary tube at both locations It implies that the straight capillary tube has higher magnetic permeability than the flexible capillary tube. T he pressure of T ool C might have not been high enough to show the relative motion against the tube surface needed for finishing in the case of straight tubes but sufficient for the flexible tubes.
130 CHAPTER 7 DISCUSSIONS AND CONC LUSIONS 7.1 Fundamental Finishing Characteristi cs of Flexible Capillary Tubes The application of magnetic abrasive finishing to the deburring of flexible capillary tubes with multiple laser machined slits has been examined and the conditions to achieve the internal deburring based on the finishing expe riments have been studied. T he finishing experiments showed the feasibility of magnetic abrasive finishing for both surface and edge finishing of the flexible capillary tub es H ard sharp cutting edges such as those of diamond abrasive are necessary for re moving laser machined burrs. Maintaining the presence of nonferrous diamond abrasive at the finishing area is a key factor to deburring the inside of the flex ible capillary tubes. In order to keep the nonferrous abrasive at the finishing area, t he ferrous particles must have microasperities on their surfaces Additionally t he diamond abrasive should be larger than the slit width to avoid losing the abrasive cutting edges from the finishing area. T he viscosity of the lubricant should be high enough to avoid leaking through the slits and to encourage the lubrication between the abrasive cutting edges and target surface but low enough to be introduced into the capillary tube. The role of the large ferrous tools mixed with the magnetic abrasive is to enhance th e finishing force by increasing the magnetic force acting on the magnetic abrasive inside the capillary tube and thereby achiev ing the desired finishing behavior. In particular, the iron particles are able to form chains that conform to the tube interior l ike a flexible brush T h e flexibility of the chains facilitates surface fine finishing. The ferromagnetic rod (carbon steel rod) generates superior magnetic force because of its high susceptibility and large volume. This pushes magnetic abrasive rigidly ag ainst the
131 target surface, enhancing the material removal for deburring. T h e ferromagnetic rod with magnetic anisotropy ( 304 stainless steel rod) develops weak er magnetic force and push es the magnetic abrasive with only one end. This causes u nstable rod mot ion and reduces the finishing capability. The shape of the ferrous rod maintains the straightness of the flexible tube during internal finishing and its increased volume enhance s the magnetic force acting on the magnetic abrasive. The combination of the ro d and iron particles in a hybrid method was consider ed and it was successful in achiev ing fine finishing of both surface and edges. In the case of internal deburring of flexible capillary tubes with slits, the burrs are obstacles for the ferrous particles mixed with diamond abrasive when the particles follow the pole movement in the axial direction ; this limited the increase of the axial feed rate. As a result, the processing speed can be rather low In order to improve the processing rate, a new method is propose d to have multiple finishing locations in a single tube to finish multiple regions simultaneously. 7.2 A New Finishing Method: Multiple Pole tip Finishing System New finishing method called a multiple pole tip finishing system has been designed. A condition required to realize the multiple pole tip system is the use of a special metastable austenitic stainless steel tool with alternating magnetic and non magnetic regions This unique magnetic property facilitates simultaneous finishing of multiple sections with a short pole stroke. Th e unique tool can be simply fabricated by the partial heat treatment of 304 stainless steel tool. The XRD analysis of the tool surface revealed that the untreated section has both bcc (due to pre finishing process) an d fcc structures while the heat treated section has f cc structure only.
132 The length of the magnetic region of the tool must be such that the borders of the magnetic region of the tool correspond to the pole tip edges. Generating a short magnetic region on t he tool with edges corresponding to the pole tip edges concentrates the magnetic flux at the pole tip edges which doubles the number of borders of the magnetic regions to attract the magnetic abrasive. T he use of this method attract s the magnetic abrasive more strongly to the tool by magnetic force and improves the deliverability of the magnetic abrasive to desired area s deeper in the tube. This achieves the uniform surface finishing in the entire finishing area. The insertion of a magnetic tool with magne tic abrasive facilitates the remov al of material from the peaks of the surface asperities by the magnetic abrasive when lodged in between the tool and target surface. This results in a smoothly finished surface with less material removal than the use of ir on particle s only ( used in the single pole tip system ) Moreover, a method was proposed to define the pole tip feed length that sufficiently achieves a uniform desired surface roughness on the entire target surface by calculating pole tip coverage time ove r the target surface. 7.3 Development of High speed Internal Finishing Machine In order to improve the processing rate, a high speed finishing system has been developed for finishing capillar y tubes and finishing experiments have been performed with tub e revolutions up to 30000 min 1 The machine demonstrate s its finishing performance successfully in comparison with the previous high speed machine in Japan and existing low speed machine. This machine improved the processing rate by a factor of 5.4 when c ompared to the low speed finishing machine with single pole tip system.
133 7.4 Internal Finishing of Capillary Tubes by High speed Multiple Pole tip System: Straight Capillaries A multiple pole tip system has been applied to the high speed finishing machine for finishing capillar y tubes and finishing experiments have been performed with tube revolutions up to 30000 min 1 In the single pole tip system, the magnetic abrasive is stable and performs efficient surface finishing up to 30000 min 1 Conversely th e magnetic abrasive and tool lapse into unstable conditions in the multiple pole tip system at high speed due to high centrifugal force. This causes deep scratches and irregular aspe rities on the finished surface. Moreover, lack of lubricant disturbs the r elative motion of abrasives against the inner surface and makes the abrasives adhere on the surface due to heat generation Increased machining rate in the system consumes more lubricant and requires fresh lubrication more frequently. The high speed multip le pole tip system has been constructed and successfully achieves surface finishing up to a tube revolution of 10000 min 1 It produces a smoothly finished surface ( 0.1 m Rz ) and is twelve times more efficient than the previous si ngle pole tip finishing system. The increase of the tube rotational speed facilitates the removal of material from the peaks of surface asperities. The material removal per tube rotation is decreased with increasing tube rotational speed. 7. 5 Internal Finishing of Capillary Tubes by High speed Multiple Pole tip System: Flexible Capillaries Based on the results from straight capillary tube finishing, the flexible capillary tubes have been ap plied for internal deburring process using each partially heat treated magnetic tool, Tool C and D.
134 Although the T ool C has i nstability during the tube rotation, with stronger magnetic field, this condition assisted the tool and particles to follow the pol e reciprocating motion and ke ep the abrasives at the finishing area Stronger magnetic force of the T ool C and its powder distribution allow the successful internal surface and edge finishing of flexible capillary tube. Conversely, ev en though the Tool D achieves successful internal finishing in straight capillary tubes, this tool re t ains the burrs and re solidified material after finishing process in flexible capillary tubes because of weak magnetic field in the finishing area T h e powder distribution in T ool D increase s the chance for particles clogging and particle loss during the pole reciprocation motion due to its weaker magnetic force This condition results in a slower processing rate when compared with T ool C Effects of tube magnetism have been st udied with finite element method and magnetic flux density measurement. Thinner tube (straight capillary) is almost magnetically saturated and has stronger magnetic field in the finishing area. In the other hand, in the case of thicker tube (flexible capil lary), the magnetic flux flows into the tube and has weaker magnetic field in the finishing area.
135 CHAPTER 8 REMAINING WORK 8.1 Development of Higher Numbered Pole tip System The development of a multiple pole tip system using a partially heat treated ma gnetic tool allows the finishing of multiple regions simultaneously in capillary tubes and improves the finishing efficiency. Although the increase of finishing efficiency in double pole tip system has been achiev ed, more number of pole tip in system will be require d for longer capillary tubes (length of catheter shaft: up to 1 m) Increase of pole tip set will enhance the finishing efficiency in capillary tube finishing and allows the longer tube finishing The configuration of pole tip affect s the intensi ty of magnetic field at the finishing area, therefore the effects of this geometry and arrangement on the magnetic field should be studied. As the tube length is getting longer, it is difficult to introduce the abrasive deep into the capillary tube. To av oid that situation, a long tool embedded abrasives on the ferrous area needs to be fabricated E ven t hough the lubricant can be easily leaked through the slits during the workpiece rotation i ncreased tube rotational speed and the number of finishing spots will consume the injected lubricant more quickly which entails the magnetic abrasive to have a higher tendency to adhere on the surface due to the heat generation from frictional forces This occurrence critically affects the behavior of the magnetic to ol and magnetic abrasives in high speed multiple pole tip finishing system. T he method related to lubrication circulating system for capillary tube is required for further study.
136 8.2 Pole Rotation System for Flexible Capillary Tubes Due to the long lengt h of flexible capillary tubes (up to 1 m) ; it is difficult to rotate the tube at high speed because of the t orsion acting on the tube even though multi numbered pole tips support the tube A pole rotation system m ight be a lternated to solve this problem I n this case, the lubrication circulating system should be easily considerable when compared with workpiece rotation system. 8.2 Magnetic Field Analysis of Multiple Pole tip S ystem Since the magnetic property is not clearly determined in plastic ally deform ed 304 stainless steel, the magnetic field analysis has not been accomplish ed. This study should be a key to help clarify each characteristic of tool behavior in multiple pole tip system.
137 LIST OF REFERENCES 1. S. Lin, P. A. Kew, K. Cornwell, 2001, Two phase Heat Transfer to a Refrigerant in a 1 mm diameter Tube, International Journal of Refrigeration, Vol. 24:51 56. 2. R. Ankur, C. Animesh, E. Chhaya, K. Devesh, 2004, Development and Assessment of 316LVM Cardiovascular Stents, Material Science and Engineering:A Vol. 386(1 2):331 343. 3. K. Nagayoshi, K. Tsuneo, 2012, Laser and Electrochemical Complex Machining of Micro stent with On machine Three dimensional Measurement, Optics and Laser in Engineering, Vol. 30(3):354 358 4. E. A. Trillo, L. E. Murr, 1998, Effects of Carbon Content, Deformation, and Interfacial Energetic on Carbide Precipitation and Corrosion Sensitization in 304 Stainless Steel, Acta Materialia, Vol. 47(1):235 245. 5. N. B. Dahotre, S. P. Harimkar, 2008, Laser Fabrication and M achining of Materials, Spr inger, NY. 6. J. Meijer, K. Du, A. Gillner, D. Hoffmann, V. S. Kovalenko, T. Masuzawa, A. Ostendorf, R. Poprawe, W. Schulz 2002, Laser Machining by S hort and U ltrashort P ulses, S tate of the A rt and N ew O pportunities in the A ge of the P hotons CIRP Annals M anufacturing Technology, V ol. 5 1 ( 2 ) : 531 550. 7. K. Enda, G. Vlado, 2009, Catheter and Specialty Needle Alloy, Materials & Processes for Medical Devices Conference & Exposition, Minneapolis, August 10 12. 8. Y. P. Kathuria, 2005, Laser Microprocessing of Metallic Stents for Medical Therapy, Journal of Materials Processing Technology, Vol. 170(3):545 550. 9. Y. P. Kathuria, 1997, Laser Precision Processing in Microtechnology. Part1 Nd YAG laser processing Proceedings of LANE 97, Germany, 272. 10. K. D. Avanish, Y. Vinod, 2008, Experimental Study of Nd:YAG Laser Beam Machining An Overview, Journal of Materials Processing Technology, Vol. 195(1 3):15 26. 11. T. Whelan, S. Parmelee, 2009, Micro Abrasive Blasting Delivers 21st Century Stents, Med ical Device Technol ogies Vol. 2 0(5):30 1. 12. H. Zhao, J. V. Humbeeck, S. Jrgen, I. D. Scheerder, 2002, Electrochemical Polishing of 316L Stainless Steel Slotted Tube Coronary Stents, Journal of Materials Science, Materials in Medicine, Vol. 13(10):911 916.
138 13. J. Shih, 2008, Biomedical Manuf acturing: A New Frontier of Manufacturing Research, Journal of Manufacturing Science and Engineering, Vol. 130( 2):919 926. 14. H. Yamaguchi, J. Kang, 2010 Study of Internal Deburring of Capillary Tubes with Multiple Laser machined Slits, Burrs Analysis, Contr ol and Removal, Part 7: 205 212. 15. E. S. Lee, 2000, Machining Characteristics of the Electropolishing of Stainless Steel (STS316L), Journal of A dvanced M anufacturing T echnology, Vol. 16(8):591 599. 16. Z. E. Geller, K. Albrecht, J. Dobranszky, 2008, Electropolis hing of Coronary Stents, Material Science Forum, Vol. 589:367 372. 17. W. C. Elmore, 1939, Electrolytic Polishing, Journal of Applied Physics, Vol. 10(10):724 727. 18. L. K. Gillespie, 2006, Mass Finishing Handbook, Industrial Press, Inc., New York. 19. J. P. Caire, E. B. Chainet, N. P. Valenti, 1993, Study of A New Stainless Steel Electropolishing Process, Proceedings of the 80 th AESF Annual Technical Conference in USA, 149 156. 20. S Mohan, D. Kanagraraj, R. Sindhuja, S. Vijayalakshmi, N. G. Renganathan, 2001, Electro polishing of Stainless Steel: A Review, Transactions of the Institute of Metal Finishing, Vol. 79(4):140 142. 20 K. M. D Silva, H. S. J. Altena, J. A. McGeough, 2003, Influence of Electrolyte Concentration on Copying Accuracy of Precision ECM, CIRP Annals M a nufacturing Technology, Vol. 52 ( 1 ) :65 168. 21 H. Z Choi, S. W. Lee, Y. J. Choi, G. H. Kim, 2004, Micro Deburring Technology using Ultrasonic Vibration with Abrasive, Proceedings of the 7th International Conference on Deburring and Surface Finishing, 231 23 8. 22 H. Z Choi, S. W. Lee, Y. J. Choi, S. L. Ko 2003, Micro Deburring Technology using Ultrasonic Cavitation, Proceedings of International Conference on Leading Edge Manufacturing In 21st Century : LEM21, 1013 1018. 23 Thermal Deburring Apparatus and Method, US Patent 4561839. 24 T. K. Brockbank, 2004, Thermal Energy Deburring, Proceedings of the 7th International Conference on Deburring and Surface Finishing, 225 230. 25 D. P. DeLo, J. M. Greenslet, G. Munko, 2007, Improved Microhole Precision and Performance by Mi croFlow Abrasive Flow Machining, Proceedings of the 15th International Symposium on Electromachining, 399 403.
139 26 S. Sehijpal, H. S. Shan, 2002, Development of Magnetic Abrasive Flow Machining Process, International Journal of Machine Tools and Manufacture, V ol. 42(8):953 959. 27 J. D. Kim, Y. H. Kang, Y. H. Bae, S. W. Lee, 1997, Development of A Magnetic Abrasive Jet Machining System for Precision Internal Polishing of Circular Tubes, Journal of Materials Processing Technology, Vol. 71(3):384 393. 28 S. H. Lee, D A. Dornfeld, 2001, Precision Laser Deburring, ASME Journal of Man ufacturing Science and Engineering, Vol. 123 : 601 608. 29 O. Sugiura, N. Imahashi, M. Mizuguchi, 1997, Development of A C ylindrical T ype M agnetic B arrel F inishing M achine, Journal of the Japan Society for Precision Engineering, Vol. 63 ( 3 ): 399 403. 30 S.L. Ko, Y.M. Baron, J.I. Park, 2007, Micro Deburring for Precision Parts using Magnetic Abrasive Finishing Method, Journal of Materials Processing Technology, Vol. 187 188 : 19 25. 31 T. Shinmura, 1990, St udy on Magnetic Abrasive Finishing, CIRP Annals Manufacturing Technology, Vol. 39(1): 325 328. 32 S. Yin, T. Shinmura, 2004, Vertical V ibration assisted M agnetic A brasive F inishing and D eburring for M agnesium A lloy, International Journal of Machine Tools & Manufacture, Vol. 44 ( 12 13 ): 1297 1303. 33 H. Yamaguchi, J. Kang, 2010, Study of Ferrous Tools in Internal Surface and Edge Finishing of Flexible Capillary Tubes by Magnetic Abrasive Finishing, Transactions of NAMRI/SME, Vol. 38:177 184. 34 H. Yamaguchi, T. Shinm ura, R. Ikeda, 2007, Study of Internal Finishing of Austenitic Stainless Steel Capillary Tubes by Magnetic Abrasive Finishing. ASME Journal of Manufacturing Science and Engineering. Vol. 129(5):885 892. 35 J. Kang, H. Yamaguchi, 2010, Internal Finishing of Ca pillary Tubes by Magnetic Abrasive Finishing using a Metastable Austenitic Stainless Steel Tool. Proceedings of the Twenty fourth Annual ASPE Meeting, 1 4. 36 J. Svoboda, 2004, Magnetic Techniques for the Treatment of Materials Kluwer Academic 37 H. Yamaguchi, J. Kang, F. Hashimoto, 2011, Metastable Austenitic Stainless Steel Tool for Magnetic Abrasive Finishing. CIRP Annals Manufacturing Technology Vol. 60(1):339 342. 38 J. Kang, H. Yamaguchi, 2012, Internal Finishing of Capillary Tubes by Magnetic Abrasive Fini shing using a Multiple Pole tip System, Precision Engineering, Vol. 36(3):510 516.
140 39 F. W. Preston, 1927, The Theory and Design of Plate Glass Polishing Machines, Journal of the Soc iety of Glass Technology, Vol. 11, 214 256. 40 Y. M. Baron, 1975, Technology of Abrasive Finishing in Magnetic Field, Mashinostroenie, Leningrad. 41 H. Yamaguchi, T. Shinmura, T. Kaneko, 1996, Development of A New Internal Finishing Process Applying Magnetic Abrasive Finishing by Use of Pole Rotation System, Int ernational Journal of the Japan. Soc iety for Precision Engineering Vol. 30 ( 4 ): 317 322. 42 T. Shinmura, Y Hamano H. Yamaguchi, 1998 A New Precision Deburring Process for Inside Tubes by the Application of Magnetic Abrasive Machining, 1st Report, On the Process Principle and a Few D eburring Characteristics (in Japanese) Transaction of the Japan Society of Mechanical Engineers ., Vol. 64 ( 620 ) Series C, 1428 1434 43 S. L. Ko, Y. M. Baron, J. W. Chae, V. S. Polishuk, 2003, Development of Deburring Technology for Drilling Burrs Using Magn etic Abrasive Finishing Method. Proceedings of International Conference on Leading Edge Manufacturing in 21st Century, Japan Society of Mechanical Engineers, 367 372. 44 K. Suzuki, H. Takahashi, H. Ohashi, T. Yonemura, E. Miyasaki, T. Uematsu, 1990, The Resea rch on Magnetic Grinding Method using Short Fiber Magnetic Abrasives 1st Report, The Effect of the Metal Short Fiber Contamination to Magnetic Abrasives. Proceedings of the JSPE Spring Annual Meeting, 307 308 [in Japanese]. 45 T. Sato, H. Yamaguchi, T. Shinm ura, T. Okazaki, 2006, Study of Surface Finishing Process using Magneto rheological Fluid (MRF) 2nd report: Effects of the Finishing Behavior of MRF based Slurry on Finishing Characteristics Journal of Japan Society for Precision Engineering, Vol. 72(11 ):1402 1406 [in Japanese]. 46 T. Mori, K. Hikota, Y. Kawashima, 2003, Clarification of Magnetic Abrasive Finishing Mechanism, Journal of Material Processing Technology. Vol. 143 144(20):682 686. 47 H. Yamaguchi, T. Shinmura, M. Takenaga 2003, Development of a N ew Precision Internal Machining Process Using an Alternating Magnetic Field, Precision Engineering, Vol. 27 ( 1 ): 51 58. 48 H. Yamaguchi, T. Shinmura, M. Sekine, 2005, Uniform Internal Finishing of SUS304 Stainless Steel Bent Tube Using a Magnetic Abrasive Fini shing Process, ASME Journal of Manufacturing Science and Engineering Vol. 127(3):605 611
141 49 H. Yamaguchi, T. Shinmura, 1995, Study on a New Internal Finishing Process by the Application of Magnetic Abrasive Machining 4th Report, Effects of Diameter of Magn etic Abrasives on Finishing Characteristics Transactions of the Japan Society of Mechanical Engineers Series C, Vol. 61(591):4470 4475 [in Japanese]. 50 H. Yamaguchi, K. Hanada, 2008, Development of Spherical Magnetic Abrasive Made by plasma Spray, ASME Jou rnal of Manufacturing Science and Engineering, Vol. 130(3):999 1008. 51 T. Shinmura, H. Yamaguchi, 1995, Study on a New Internal Finishing Process by the Application of Magnetic Abrasive Machining Internal Finishing of Stainless Steel Tube and Clean Gas Bomb International Journal of Japan Society of Mechanical Engineering Series C, Vol. 38(4):798 804. 52 T. Shinmura, H. Yamaguchi, T. Aizawa, 1993, A New Internal Finishing Process of Non ferromagnetic Tubing by the Application of a Magnetic Field The Developmen t of a Unit Type Finishing Apparatus using Permanent Magnets, International Journal of Japan Society for Precision Engineering, Vol. 27(2):132 137. 53 Y. H. Zou, T. Shinmura, 2009, A New Internal Magnetic Field Assisted Machining Process using a Magnetic Mach ining Jig Machining Characteristics of Inside Finishing of a SUS304 Stainless Steel Tube, Advanced Material Research, Vol. 69 70:143 147. 54 H. Kawakubo, K. Tsuchiya, T. Shinmura, 2003, Consideration on Mechanism of Polishing Force in Magnetic Polishing Meth od using Loose Abrasive, Journal of Japan Society for Abrasive Technologies, Vol. 47(12):672 676 [in Japanese]. 55 H. Nishida, K. Shimada, Y. Ido, 2008, Fluid Dynamic Investigation of Polishing the Inner Wall of a Tube Utilizing a Magnetic Compound Fluid (MCF ), Proceedings of the 7th JFPS International Symposium on Fluid Power, TOYAMA, 837 840. 56 W. Kordonski, D. Go lini, 2002, Multiple Application of Magnetorheological Effect in High Precision Finishing, Journal of Intelligent Material Systems and Structures, Vol. 13:401 404. 57 H. Yamaguchi, T. Shinmura, S. Watanabe, 2001, Study of a Magnetic Field Assisted Internal Fi nishing Process using Abrasive Slurry Circulation System Effects of Geometrical and Magnetic Properties of Magnetic Particles on the Finishing Characteristics Journal of the Japan Society for Precision Engineering, Vol. 67(3):444 449[in Japanese]. 58 H. Ya maguchi, T. Shinmura, N. Horiuchi, 2001, Study of Magnetic Tool Behavior and Its Relationship to the Processing Characteristics in a Magnetic Field Assisted Barrel Finishing Process, Transactions of NAMRI/SME, Vol. 29:197 204.
142 59 K. Mumtaz, S. Takahashi, J. E chigoya, Y. Kamada, L. F. Zhang, H. Kikuchi, K. Ara, M. Sato, 2004, Magnetic Measurements of Martensitic Transformation in Austenitic Stainless Steel after Room Temperature Rolling, Journal of Materials Science, Vol. 39:85 97. 60 S. Takahashi, J. Echigoya, T. Ueda, X. Li, H. Hatafuku, 2001, Martensitic transformation due to plastic deformation and magnetic properties in SUS 304 stainless steel, Journal of Materials Processing Technology, Vol. 108:213 216. 61 P L. Mangonon G Thomas 1970, T he mart e nsite phases in 304 stainless steel Metallurgical and Materials Transactions B, Vol. 1(6):1577 1586. 62 J. Childress, S. H. Liou, C. L. Chien, 1988, Magnetic Properties of Metastable 304 Stainless Steel with BCC Structure. Journal de Physique Colloques, Vol. 49(C8):113 1 14. 63 S.S.M. Tavares, D. Fruchart, S. Miraglia, 2000, A magnetic study of the reversion Journal of Alloys and Compounds Vol. 307:311 317. 64 K&J Magnetics. Inc. catalog. http://www.kjmagnetics.com/proddetail.asp?prod=B888 November 18, 2012 (Accessed date), Tel.1 215 766 8055 65 H. Yamaguchi, T. Shinmura, 1994, Study on a New Internal Finishing Process by the Application of Magnetic Abrasive Machining ( 2nd Report, Effects of Magnetic Field Distribution on Magnetic Force Acting on Magnetic Abrasives), Transactions of the Japan Society of Mechanical Engineer s Series C Vol. 60 ( 578 ): 3539 3545. [in Japanese] 66 T. Sato, H. Yamaguchi, T. Shinmura, T. Okazaki, 2 009, Study of Internal Finishing Process for Capillary using Magneto rheological Fluid, J ournal of Japan Society of Prec ision Eng ineering Vol. 75 ( 5 ): 612 616. [in Japanese] 67 J. Kang, A. George, H. Yamaguchi, 2012, High speed Internal Finishing of Capillary Tubes by Magnetic Abrasive Finishing, Proceeding of CIRP, Vol. 1: 414 418. 68 H. Yamaguchi, T. Shinmura, 1995, Magnetic Abrasive Finishing of Inner Surface of Tubes, Proceedings of International Symposium for Eletromachining, 963 976.
143 BIOGRAPHICAL SKETCH Ju nmo was born in Guri si, Gyeonggi do, South Korea in 1980. He obtained a B.S degree in mechanical engineering from the Sungkyunkwan University (SKKU) in South Korea in 2006. He served in the military service (Korean Army) for 26 month s from 2000 to 2003 a s a field service technician At SKKU, he re c eived a scholarship and a few prizes for designing the new c oncept tiller. After graduatio n, he came to United States for his Ph.D. study and joined Dr. Hitomi Greenslet Mechanical a nd Aerospace Engineering at the University of Florida (UF) in 2008 He received an Academic Achievement Award from UF in 200 7 During his Ph.D. studies, he was sponsored by the UF Research Foundation (Gatorade) to fabricate the finishing machine, and recei ved an E. Wayne Kay Graduate Scholarship from the Society of Manufacturing Engineers (SME) Education Foundation He wishes to continue his career as a researcher.