Aalysis of Material Removal Rate ad Surface Roughess o Advaced Ceramics Materials Micro- Hole Drillig Shu-Lug Wag #, Tig-Yu Chueh #2 # Departmet of Mechaical Egieerig, Taoyua Iovatio Istitute of Techology, Taiwa #2 Departmet of Isurace ad Fiace, Natioal Taichug Uiversity of Sciece ad Techology, Taiwa wzl@tiit.edu.tw, 2 cty@utc.edu.tw DOI: 0.563/00003 Abstract-This study discussed the rotary ultrasoic machiig (RUM) parameters ad tool geometry parameters of optimized hole wall quality, ad obtaied better ad uiformly distributed surface roughess of deep hole walls. The experimetal cofiguratio of this study was plaed usig Taguchi orthogoal array, where the hole diameter, spidle speed, feed rate ad power were take as the experimetal factors. Drillig machiig experimet was carried out usig the parameter combiatio cofigured by the experimetal desig, ad wall surface roughess of the drilled holes was measured. A relatio model betwee the average surface roughess ad stadard deviatio of the hole walls ad processig parameters ad tool geometry were established, so as to achieve optimized surface roughess of the overall deep hole walls. The optimizatio process respectively aimed at the miimum surface roughess ad maximum metal removal rate. The experimetal results proved that applicatio of the optimized parameter combiatio i rotary ultrasoic machiig of hole processig ca obtai better ad more uiform hole wall surface quality.the effect of cotrol factors o the hole quality was discussed from the respects of the straightess deviatio, miimum surface roughess ad maximum metal removal rate of related factors o the hole whe the Taguchi method was applied to rotary ultrasoic machiig of hole drillig. The results showed that the level of straightess factors of produced holes ca be fully preseted. Keywords-Advaced Ceramics; Rotary Ultrasoic Machiig; Surface Roughess; Material Removal Rate; Holes Deviatio; Taguchi Method I. INTRODUCTION The Istitute for Productio Techology of Uiversity of Tokyo used Ultra Soic Machiig (USM) to drill 5μm-dia. micro-holes o hard brittle materials like glass, silica gel ad ceramics. Wire Electrical Discharge Machiig (WEDM) was first used to develop micro cylidrical tool o machiig apparatus; the with machiig apparatus workig, μmamplitude ultrasoic vibratio occurred o the side of machiig object; fially sub-micro diamod abrasive grais were utilized to cotrol processig force. Therefore, the method ca be used for micro hole drillig. WEDM is a method which may be applied to make micro tools, ad eables such tools made o machiig apparatus to adjust their cetral poits automatically, so extremely low possibility of eccetricity could be esured. The method is widely applied. Besides circular hole drillig ad irregularly-shaped hole, it ca also be employed for 3D shaped hole makig. I recet years, with idustry maufacturig techology gradually advacig as product precisio cotiually improvig, more ad more attetio was attracted to maagemet o product surface roughess i productio site, ispectio area, stadard testig laboratory, research laboratory, etc. However, abrasio ad lubricatio caused by surface characteristics of work pieces i service, adhesive capacity of compositio surface of work pieces, air-tightess ad oil-tightess of slidig surface of work pieces, ad surface fiish exert direct ad importat iflueces o pursuig higher quality ad better stability of the products i the idustry. Take mechaical fuctios of work pieces for example, their surface structures are closely correlated with their performace; especially whe work pieces are i cotact with the surface of other work pieces, better smoothess is preferable; sometimes, however, some roughess is required. Therefore, to cofirm mechaical performace of work pieces, surface roughess must be measured. Cuttig process is a extesive maufacturig process i the idustry. To improve product quality is absolutely oe of the goals for all kids of cuttig operatios, ad machied surface roughess is also amog importat research topics relatig to the quality of maufacturig process. The vibratio of cuttig tool exerts importat effect o hole shape. As for the researches i the respect, Sakuma et al. [] (0) foud that vibratio frequecy was correlated with the umber of machied hole corers by aalysig vibratio frequecy of cuttig tool. Gessesse et al. [2] carried out processig tests of BTA deep holes uder differet spidle speeds, ad observed the effect of processig parameters such as abrasio status of work piece material ad cuttig tool o spiral outlie of hole shape. I 2, Katsuki et al. [3] used gu drill ad log twist drill to process work pieces with iclied plae, ad discussed hole deviatio. Experimetal results showed that cuttig force directio might affect hole deviatio o iclied plae of work pieces; hole deviatio caused by gu drill may go towards positive directio of horizotal axis ad that caused by log twist drill towards egative positio of horizotal axis. I 0, Ei-Khabeery et al. [4] used experimetal approaches ad processig parameters like rotatio speed, feed rate ad hole diameter for processig experimet of deep holes with gu drill, ad utilized surface roughess of hole wall o discuss the ifluece of rotatio speed, feed rate ad hole diameter. Experimetal results showed that better surface roughess of hole wall could be achieved with high rotatio speed, low feed rate ad small hole diameter. Although lots of scholars have researched processig of deep holes ad quality of hole wall [4], [5], [6], these researches dealt with the quality of partial hole wall rather tha whole hole wall. This research maily uses the experimetal method to discuss the ifluece of RUSM o the whole quality of deep Vol. No. 202 PP.-5 www.jphf.org c World Academic Publishig - -
hole i ceramics materials. I the process of machiig, brittle failure (Micro scraps) is still domiatig the removal mechaism of ceramics materials, if the machiig parameters ca t be cotrolled properly, it will cause poor roughess of the deeper subsurface crack. The aalysis of this research cosists of three parts: firstly, it aalyses the itegrity of ceramics materials i the process of RUM ad the correlatios betwee RUM ad machiig coditios through the results of axial sub-sectioal measure i meas ad stadard deviatios of statistics; secodly, it discusses the machiig parameters such as tool diameter, ultrasoic power desity, speed of mai shaft ad feedig amout which are the keys to the research i the coditio that grai size is fixed; fially, it discusses the iflueces of Taguchi quality experimetal desig method o machiig quality with the expectatio achievig better ad more uiform surface quality of hole wall to esure the hole quality after drillig ad improve the machiig rate. experimetal desig ad coditios A. Experimetal Desig ad Coditios: This experimet is a precisio test of RUM machiig. Besides improvig RUM techology ad cotrollig the tolerace stably uder 0.0mm, it also compares the precisio testig methods of projectio-type, optical microscopy true roudess, ad profile testig method with each other. The experimetal machie of RUM is of UMT-6 triaxial CNC ultrasoic wave, while the work piece materials is of ceramic glass (Zerodur), amely α (coefficiet of thermal expasio of zero) of advaced glass material of which the physical property is showed i Table. This research iitially adopts the Taguchi aalysis ad the cosistetly discusses the optimal profilig. RUM precisio processig test is costructed by usig rotary super ultrasoic processig machie for effects show from mai measuremet tests icludig hole roughess, o the basis of the o. of fixed particle i diamod kife (#320) ad the diameter of diamod kife (36μ~54μ), so four processig parameters are ecessarily put ito plas, which are kife diameter, machie power, mai axle speed ad feed rate. as: A measuremet legth L is take from curve surface of workig piece where surface roughess exists, ad the average height of legth L is set up as x axle of cetral lie, ad vertical directio X is y axle, see i fig.. The average roughess of cetral lie is expressed by equatio: cetral lie Fig. Average roughess diagram of cetral lie () TABLE PHYSICAL FEATURES OF CERAMIC GLASS Desity:2.53g/cm3 Heat capacity:0.0j/gk Youg s modulus:0.3gpa Koop hardess:620 B. Defiitio of Hole Precisio ad Testig Methods The hole precisio ca be divided ito geometric diameters of true roughess, true straightess ad parallelism of cylidricity tolerace ad surface roughess. I additio, it also cotais relative positio tolerace of hole ad other stadard geometric shapes as plae ad colum, ad true positio degrees. The istructios of precisio measuremet istrumets are as follows: () Hommelwerke test T surface fiish measurig istrumets: Amog the methods of expressig surface roughess, the average roughess i cetral lie Ra is used most. This study will express surface roughess of hole wall by Ra value. The average roughess value i cetral lie is used most amog all expressio methods of roughess, which is defied Fig. 2 Experimetal flow chart of maufacturig steps (2) Two-Coordiate Measurig Microscope (ZKM 0-250C): The mai fuctios are to measure rectagular coordiates, polar coordiates, ad colum coordiates, ad it also ca be used i the distace measuremet betwee two surfaces of plae ad curve, ad poit, agle ad axis. It is also used to measure the straightess, parallelism ad square trowel. (3) Calculatio method of material removal rate: Accordig to [2] the followig equatio is used to calculate the material removal rate by diameter of hole after drillig, drill diameter ad the time of drill puchig through the work piece(d) MRR= [π [ (Dout/2)2-(Di/2)2] d] T Where Dout: the diameter of hole after drillig Vol. No. 202 PP.-5 www.jphf.org c World Academic Publishig - 2 -
Di: the drill diameter d: the thickess of work piece (40 mm ) T: the time of drill puchig through the work piece thickess (d) TABLE 2 EXPERIMENTAL CONFIGURATION Code Factor Name Level Level 2 Level 3 A Diameter of drill bit (mm).5 2.3 B Rotatio speed (RPM) 4000 4500 5000 C Power 20 2 D Feed rate 0.4.6 C. Desig Variable of Taguchi Method ad Level Meter I drillig, hole quality is geerally decided by surface roughess. The experimet uses Taguchi Method to pla the coditios for cuttig experimet, ad applies RUSM as cuttig tool. The experimet is doe i three steps: first, four experimetal parameters are chose based o surface roughess affectig cuttig operatio, i.e. diameter of drill bit, rotatio speed, feed rate ad power, as show i Table 2; secod, three level values are give to each experimetal parameter, ad groups of tests are carried out accordig to Orthogoal Array (OA) of L (32) chose; third, mea value of Taguchi Method is used for aalysis, ad the characteristics of surface roughess of target fuctio are coverted ito Sigal Noise Ratio (S/N). S / N = 0 log ( 2 0 y i i= where S deotes stadard deviatio, which is defied as follows: Experimetal parameters are assessed accordig to the formulae above, ad the parameter with greatest effect o surface roughess is diameter of drill bit, followed by rotatio speed, feed rate ad power. The experimet is desiged to obtai average ad correctess ad plaed ito OA of L (32), totally with groups of experimets ivolved, so as to improve experimetal accuracy. Experimetal flow chart of maufacturig steps is give i Fig. 2, ad desig steps of experimetal parameters i Fig. 3. Fig. 3 Desig steps of experimetal parameters ) (2) II. RESULT AND DISCUSSION Hole drillig o ceramic material is differet from processig geeral metal materials. The quality of hole drillig, which is very importat, is geerally assessed by comparative aalysis of roughess, roudess ad parallelism etc. The process parameters affectig the roughess of hole iclude spidle speed, feed rate ad ultrasoic power. May related works have compared the ultrasoic machiig [7, ] ad rotary ultrasoic machiig, such as the experimetal data of brittle materials i respect of the mai spidle speed, with the coclusio that the material removal rate (MRR) icreased with the icrease of circumferetial speed of drillig bit [0, ]. This experimet fixes diamod tool particle umber (#320) ad the diamod particle diameter (36μ~54μ) ad varies the four machiig parameters, icludig diameter of drill, rotatio speed, power, ad feed rate, to discuss their impacts o the roughess. The diameter of drilled hole is obtaied through referrig to the data measured by surface roughess tester ad two-coordiate measurig microscope (ZKM 0-250 C) ad makig aalysis based o the followig coditios: A. Optimizatio Aalysis of Borig Parameters of RUM Accordig to the results of borig data, the relatioship ad framework of surface roughess ad borig parameters are discussed for gettig the optimum borig parameters ad framework of surface roughess combiatio o the maximum MRR ad miimal roughess of ceramic materials. The MRR experimet utilizes the LTB (larger the better) characteristics to carry out S/N ratio ad plots the results as Table 3 ad Fig. 4. I additio, the surface roughess experimet employs the STB (smaller the better) characteristics for S/N ratio ad plots the results as Table 5 ad Fig. 5. ) MRR Aalysis Accordig to the results of borig data, the relatioship is discussed for gettig the optimal borig parameter combied with the maximum MRR of ceramic materials. The MRR experimet utilizes the LTB (larger the better) characteristics to carry out S/N ratio ad plots the results as Table 3 ad Fig. 4. The cotributio o MRR of every factor is show i Table 4. For borig ad machiig of ceramic materials as preseted i Table 3, with drill diameter, rotatio speed, power ad feed rate of differet cuttigs, as well as fixed diamod tool particle umber (#320) ad diamod particle diameter (36μ~54μ), if the diameter is the same, the larger the rotatio speed, power ad feed rate are, the bigger the MRR is. As preseted i Table 3, there is a large MRR (0.54 mm3/mi) with the drill diameter (ψ.5), the rotatio speed (5000RPM), the power (2) ad the feed rate (.6). For the same drill diameter (ψ.5), the sub-rotatio speed (4500RPM), the sub-power (20) ad the sub-feed rate (.0), the MRR is still large(0.2mm3/mi). Thus, for a fixed diameter of drill, the large rotatio speed, power ad feed rate still have a large MRR, as show i Table 3. Aalysis of variace (ANOVA) ca give more objective judgmet o the relative effects betwee differet factors. The table 4 is the ANOVA table for this process. The larger the F value is, the more importat the factor for the MRR is. From Table 4 we kow, the cotributios of the factors are i the order as: rotatio speed>diameter of drill >feed rate> power. Vol. No. 202 PP.-5 www.jphf.org c World Academic Publishig - 3 -
The reaso is stated as below, the diamod particles of tools with a fixed drill diameter are the same, so whe the MRR is cosidered oly, i the scope of this experimetal cofiguratio see as Table 2, the icrease of rotatio speed provides higher probability for diamod particle to cotact, grid ad machiig workpieces, ad cosequetly the impact of rotatio speed is higher tha that of other parameters (diameter of drill, feed rate ad power). Therefore, we obtai the optimum machiig parameter ABCD3 preseted as Fig. 4. This idicates that diameter of drill (ψ), rotatio speed (4000), power () ad feed rate (.6) ca get the optimum MRR. Sequece A B C D Diameter of Drill Bit TABLE 3 MRR AUXILIARY TABLE Rotatio Speed Power Feed Rate MRR (mm 3 /mi) 4000 0.4 0.002 2 4500 20.0 0.033 3 5000 2.6 0.055 4.5 4000 0.4 0.025 5.5 4500 20.0 0.2 6.5 5000 2.6 0.54 7 2.3 4000 0.4 0.04455 2.3 4500 20.0 0.0202 2.3 5000 2.6 0.07532 TABLE 4 ANOVA TABLE Factors 2 3 DO F SS SMS F Cotributio Diamete 30. 54.5 77.2 4.004763 r 3.0 22.23663 6 2 4 7 3 26.465 Rotatio Speed 36.65 27.003200 20.6 3 2 30. 3 5. 4 0.27 4 66.27 Power 2.27 2.0626 5 26. 6 2.02 3 4.0 0.2075 2.3736 Feed Rate 27.0 26.50704 30.6 2 30.56 5.2 0.7240 5.2347 2) Roughess Aalysis Accordig to the results of borig data, the relatioship ad framework o surface roughess ad borig parameters are discussed for gettig the optimum borig parameter combied with the miimum roughess of ceramic materials. The surface roughess experimet employs the STB (smaller the better) characteristics for S/N ratio ad plots the results as show i Table 5 ad Fig.5. The cotributio of every factor o roughess is show i Table 6. This research deals with borig ad machiig of ceramic materials, with drill diameter, rotatio speed, power ad feed rate of differet cuttigs, as well as fixed diamod tool particle umber (#320) ad diamod particle diameter (36μ~54μ), if the diameter is the same, the larger the feed rate, power ad rotatio speed are, the larger the roughess is. As preseted i Table 5, there is a small roughess (0.5μm) with the drill diameter (ψ.5), the rotatio speed (5000RPM), the power (2) ad the feed rate (.6). Or with the differet drill diameters (ψ.0, ψ.5) ad the same rotatio speed (4000RPM), power () ad feed rate (0.4), the roughess is sub-low (0.μm). Thus, for a fixed diameter of drill (ψ.5), the larger feed rate, power ad rotatio speed have still a lower roughess, as show i Table 5. 37 32 27 22 7 2 A 42 A2 A3 B B2 B3 7 0 2 4 6 0 2 4 6 Fig 4.Optimum respose diagram TABLE 5 ROUGHNESS AUXILIARY TABLE A B C D Measured Sequece Diameter of Rotatio Data Power Feed Rate Drill Bit Speed μm 4000 0.4 0. 2 4500 20 0.2 3 5000 2.6 0.2 4.5 4000 0.4 0. 5.5 4500 20 0.27 6.5 5000 2.6 0.5 7 2.3 4000 0.4 0.24 2.3 4500 20 0.23 2.3 5000 2.6 0.24 TABLE 6 ANOVA TABLE Factors 2 3 DOF SS SMS F C Cotributio Diameter 4.24 4.24 2.52 2 6.0 3.055 0.65 2.2 Rotatio Speed 4.062 2.706 4.2 2 4.30 2.0 0.44 2.45 Power 4.73 3.757 2.5 2 6.30 3.45 0.6 33.45 Feed Rate 2. 4.24 3. 2 3.0.54 0.32 5.7 As for the aalysis of variace (ANOVA), we ca obtai a objective judgmet about the relative effect of differet factors to roughess. As show i Table 6, it is the ANOVA about the relative effect of differet factors to roughess. I additio, it usually employs F statistic ( F = s 2 /s 2 2 ) to deote the relatio of factors effect o error deviatio. The larger F statistic is, the more importat to the effect of roughess the factor is. The results ca obtai from the Table 6. To thik of the cotributio, the factors order of importace o roughess is i the order as: power> drill diameter> rotatio speed> feed rate. The reaso to cause the effect is stated as above. Uder the experimet coditio, with the differet drill diameters (ψ.0, ψ.5) ad the same rotatio speed (4000), power () ad feed rate (0.4), the roughess is sub-low(0.μm). So, we ca iduce that the lower rotatio speed, power, ad feed rate, the smaller roughess is. The diamod particles of tools with a fixed drill diameter are the same, so whe the roughess is cosidered oly, i the scope of this experimetal cofiguratio table, the icrease of rotatio speed provides higher probability for diamod particle to cotact, grid ad machiig workpieces. Also, whe the diameter of drill for borig is larger, the higher probability exists for ier hole C2 C3 D D2 D3 Vol. No. 202 PP.-5 www.jphf.org c World Academic Publishig - 4 -
wall to cotact, grid ad machiig workpieces. The probability of cotact icreases with the size of drill diameter. Cosequetly, the impacts of power, drill diameter ad rotatio speed are higher tha that of feed rate. Some experimetal results [,0,] had show that a better surface morphology ad less sub-surface damage was obtaied usig smaller grid size ad lower feed rate. That is, usig bigger size drill ad larger feed rate will cause more surface damage. Accordig to parameters cofiguratio ad experimetal results (show as Table 2 ad Table 5), with diameter of drill (ψ.5), power (2), rotatio speed(5000), ad feed rate (.6), we ca iduce that fixed diameter with larger feed rate, power ad rotatio speed ca get lower roughess(0.5μm). To discuss roughess of ceramic materials after machiig oly, we must cosider the impacts of other related parameters as well as the same diameter. I Fig. 5, we obtai the most appropriate machiig parameter A3B2C3D of the optimum respose diagram. This idicates that diameter of drill (ψ2.3), rotatio speed (4500), power (2), ad feed rate (0.4) ca get better roughess withi the experimetal cofiguratio (as show i Table 2). 6 5 4 3 2 0 A A2 A3 B B2 B3 7 0 2 4 6 0 4 6 Fig. 5 Optimum respose diagram III. CONCLUSION This paper uses RUM to establish relatioship mode of surface quality of drill hole wall o ceramic materials, ad utilizes such mode to fid out drillig parameters ad tool geometry of optimal hole wall surface quality for study o drill hole quality. The followig coclusios may be draw based o experimetal results above:.relatioal equatio of drill hole wall surface quality ad drillig parameters ad tool geometry may be used to forecast hole wall quality. To cosider the rage of experimet factors as show i Table 2, whe diameter of drill is fixed(ψ.5), we ca fid out greater MRR (0.54 mm 3 /mi) with larger rotatio speed, power ad feed rate as results show i Table 3. C C2 C3 D D D3 2.I rage of this experimet (Table 2), uder the coditios of greater rotatio speed, power ad feed rate as well as fixed diameter (ψ.5), lower roughess (0.5μm) ca be obtaied. This results show as i Table 5. 3.The observatio is carried out accordig to ifluece aalysis chart of hole deviatio, ad steady cuttig beig performed at the same itervals; i.e. uder stable actio of all parameters, optimal roughess ad greater removal rate ca be obtaied. Experimetal plaig based o OA of Taguchi Method has both advatages ad disadvatages as follows: () arragig the least umber of experimets to attai the best experimet efficiecy; (2) simple ad quick aalysis; (3) beig ot capable for aalyze multiple objective fuctio. ACKNOWLEDGMENT This study is fiacially sposored by the Natioal Sciece Coucil uder Grat No. NSC 7-222-E-253-00. REFERENCES [] Keizo Sakuma, Koichi Taguchi, ad Akio Katsuki, Study o Deep- Hole Borig by BTA System Solid Borig Tool--Behavior of Tool ad Its Effects o Profile of Machied Hole, Bull. Japa Soc. of Prec. Eg, Vol. 4 No.3 43-4, 0. [2] Y. B. Gessesse, V. N. Latiovic, ad M. O. M. Osma, O the Problem of Spirallig i BTA Deep-Hole Machiig, Joural of Egieerig for Idustry, Vol. 6 6-65, 4. [3] Akio Katsuki, Hiromichi Oikura, Keizo Sakuma, Tori Che, ad Yukitaka Murakami, The Ifluece of Workpiece Geometry o Axial Hole Deviatio i Deep Drillig, JSME Iteratioal Joural Series III, Vol. 35 No. 60-67, [4] M. M. EI-Khabeery, S.M. Saleh, ad M. R. Ramada, Some Observatios of Surface Itegrity of Deep Drillig Holes, Wear. 42 33-34, 0. [5] J. Frazao, S. Chadrashekhar, M. O. M. Osma, ad T. S. Sakar, O the Desig ad Developmet of a New BTA Tool to Icrease Productivity ad Workpiece Accuracy i Deep Hole Machiig, The Iteratioal Joural of Advaced Maufacturig Techology,. (4) 3-23, 6. [6] N. Latiovic ad M. O. M. Osma, Optimal Desig of BTA Deep- Hole Cuttig Tools with Staggered Cutters, INT. J. PROD. RES. VOL. 27 NO., 53-73,. [7] Gilmore R., Ultrasoic machiig of ceramics, SME Paper, MS0-346,0. [] Rozeberg L.D., Kazatsev V.F. et al., Ultrasoic cuttig, Cosultats Bureau, NewYork,64. [] Kohls J.B., Ultrasoic maufacturig process: Ultrasoic machiig (USM) ad ultrasoic impact gridig (USIG).The Carbide ad Tool J., 4, 6 (5), 2-5. [0] Haslehurst M., Maufacturig techology (3rd ed.),, pp.270-27. [] Soudararaja, V. ad Radhakrisha V., A experimetal ivestigatio o the basic mechaisms ivolved i ultrasoic machiig. It. J.MTDR, 6, 26 (3), 307-32. Vol. No. 202 PP.-5 www.jphf.org c World Academic Publishig - 5 -