Surface performance of workpieces processed by electrical discharge machining in gas

Journal of Harbin Institute of Technology( New Series), Vol 16, No 2, 2009Surface performance of workpieces processed by electricaldischarge machining in gasLI Li-qing, BA/ Ji-cheng, GUO Yong-feng, WANG Zhen-long李立青,白基成,郭永丰,王振龙(School of Mechatronical Technology, Harbin Institute of Technology, Harbin, 150001, China, liliqing@ hit. edu. en)Abstract The surface performance of workpieces processed by electrical discharge machining in gasEDM) was studied in this paper. Firstly, the composition, micro hardness andlaver of electricacharge machined(EDMed)surface of 45 carbon steels in air were investigated through different testmethods. The results show that the workpiece surface EDMed in air contains a certain quantity of oxideidation occurs on the workpiece surface. Compared with the surface of workpieces processed in kerosene, fewerracks exist on the dry EDMed workpiece surface, and the surface recast layer is thinner than that obtained byconventional EDM. The micro hardness of workpieces machined by dry EDM method is lower than thatchined in kerosene, and higher than that of the matrix. In addition, experiments were conducted on the surfacewear resistance of workpieces processed in air and kerosene using copper electrode and titanium alloy electrodeThe results indicate that the surface wear resistance of workpieces processed in air can be improved, and it isrelated with tool material and dielectrKey words: electrical discharge machining in gas( dry EDM); surface performance: 45 carbon steelsCLC number: TGI7Document code: AArticle ID:10059113(2009)02025505The working dielectric is another very important reported due to transformation of the workingnfluence factor besides processing power source and in traditional EDM:. However, up to now, electri-two poles in electrical discharge machining( EDMdischarge machining in gas dry EDM is still aDifferent working dielectrics produce different process-. new technology, although the researches about the ma-ing performances, which usually limited to material re- terial remove rate and tool wear of the dry EDM havemove rate, tool wear and surface integrity. For exam- been carried out, the systemic researches on the suple, the material removal rate increases if oxygen gas or face performances of dry EDM were little reportedmicro bubbles are mixed in liquid dielectric 4; the Since the surface performance is one of important comtool wear is very low in dry EDM-gases dielectricposition of the EDM processing performance, it is necplace liquid dielectric, such as de-ionized water and essary to study it to understand generally the surfacekerosene. Moreover, in dry EDM, workpiece is suit- performance in dry EDM. In this paper, dry EDMedable for serving as anode, and the tool wear almost has workpiece surface hardness, surface recast layer andno relation with the pulse duration, which is different surface wear resistance were studied compared with thefrom the traditional EDM where the tool wear has surface obtained in traditional EDM kerosene used asrelation with both the workpiece polarity and pulse du- dielectric ). At the same time, the workpiece surfaceration, for instance, the tool wear is usually high on the performances of the dry EDM were analyzed accordingfinish gauge or on the conditions of micro EDM gauge. to the research results about the mechanism of the dryFurthermore,workpiece surface performance, such as EDMsurface hardness, surface roughness and surface composition etc, can be improved or changed by mixing 1 Experimental Conditionsdifferent substance into the dielectric, usually are inor-ganic powder(aluminum powder, silicon powder etc.The samples served for beingd surfaceor organic matter(urea and sucrose)composition, surface recast layer andTherefore, the machining performance can be were processed in air with copper tool and they were 45changed due to the change of the working dielectric in carbondiffraction anaEDM. Many researches about the surface integrity were XD中国煤化工 roscope(SEM)CNMHGReceived 2007-07-17sNored by the Fund for the Doctoral Program of Higher Education(RFDP)( Grant No. CBQ024403007)and the Innovation Fund of HIT( Grant NoBQQ1840008)Journal of Harbin Institute of Technology( New Series), vol. 16, No 2, 2009analysis were adopted to study the workpiece surface Fig. I shows the surface compositions of the sample pro-composition and surface recast layer. The surface hard- cessed in kerosene, including Fe, austenite and littness test apparatus was employed to measure the sur- Fe, C not signed in Fig. 1). Fig. 2 shows that the sur-face hardness. The X-ray diffraction analytical appara- face compositions of the sample EDMed in air includetus was D/max-rBX, made in Japan, and the scanning Feo, Fe and a small quantity of austenite. Oxidation re-electronic microscope was $4700, produced by Hita- action takes place between Fe and oxygen because of thechi Company and the type of the surface hardness teg- high temperature in dry EDM, and Fe0 is formed on theting instrument was HVS-5000, made in China. In ad- workpiece surface. Energy spectrum analysis(ESA)of thedition, 45 carbon steel were machined using copper specimen machined in air further verifies the analysis andand titanium alloy(TCA)tool electrode in the medium the analysis result is shown in Tab. 2. As shown in Tabof air and kerosene to serve as the samples of the sur- 2, Fe is main ingredient. 0 and N are composing elementsface wear experiment. All the samples were processedon the HCK400 CNC machine tool, which are produceby Han Chuan machine tool Factory, China, and all the△Fetools were pipe electrodes. The machining polarity ofthe samples used for different surface testing analysiswas negative when machined in kerosene, and nofushing, while it was positive when they were pro-cessed in air medium. The open voltage was 280V forthe all the prepared samples2 Dry EDMed Workpiece Surface Performances2.1 Workpiece AppearanceThe direct influence of altering working dielectricFig 1 XDA of the sample EDMed in keroseneon workpiece is the appearance change. Different fronthe surface appearance processed in kerosene, the suface machined in gas has no metal luster, and somedebris are easily adhibited on the processed surface2000edge. In dry EDM, if the workpiece is connectex Austeniteanode of the power supply, the workpiece appearance1500shows black after it being processed in air or oxygen 8gas, while the surface show grey, just like the color ofbase stock after processed in nitrogen gas and argongas. If the workpiece is connected with cathode of e-lectric source, the work piece presents grey colorIf workpiece is connected positive polarity of thepower supply, it receives larger energy and oxidationaction occurs when it is processed in air or oxygen,therefore workpiece takes on black appearance due toFig 2 XDA of the sample EDMed in airoxide( see section 2. 3). If workpiece is machined as Tab. 1 Experimental conditions of the samples for XDAnegative polarity of the power supply, it received lessvalueheat quantity: therefore workpiece appearance takes oncolor of the base metal. The reasons why the workpiecePulse duration/μsprocessed in gas have no metal lustre are interpretedPulse interval/(1)Some debris attach on surface by gas flow;Peak current/A(2)Craters are not regular due to weakerServo reference voltage/Vpressibility of gas applying on discharge passagesTherefore, diffuse reflection is formed and resultsworkpiece surface no lustreTab 2 ESA result of the sample EDMed in air2.2 Surface Composition Analysis中国煤化工MX-ray diffraction analyses(XDA)were carried outCNMHG II8the samples EDMed in kerosene and air to realize theFe-Ksurface composition. The experimental conditions are lis-Weight precent: At: Atom nummer percentin Tab. I. Figs. 1 and 2 depict the XDA resul256Journal of Harbin Institute of Technology(New Series), Vol 16, No. 2, 20092.3 Surface Recast Layer0.024W/m·K, smaller than5W/m·K, the therScanning electron microscope SEM)analysis mo-conductivity of kerosene, which indicates the coolwas performed subject to the cross-section of the sam- ing effect of kerosene is better than air. So, the meltedples those were processed in kerosene and air on rough metal remained on the workpiece surface are cooledmachining gauge, and the processing parameters are slowly in dry EDM than that of in conventional EDMshown in Tab. 3. After been processed, the specimens which leads to smaller surface temperature grads on thewere cleaned and then the cross-sections were pol- surface obtained by dry EDM. Thus, smaller tensionished. The SEM photos of the specimen cross-section stress is generated and almost no cracks formed on theare shown in Fig. 3. We can see in Fig. 3 that the re- surface after dry EDM. The workpiece surface is heatcast layer EDMed in air is thinner than that of in kero- ed concentrated relatively in traditional EDM, espe-ene, and there are no micro cracks on the surfacecially on the condition of rough gauge Kerosene hasHowever, several micro cracks are found on the surface better cooling effect than air, then on discharge spot,machined by conventional Elthe melted metals transform into solid state very quickly. All of these result in larger temperature grads, andthat induce larger tension stress, therefore, microcracks are generated on the surface of specimenEDMed in kerosenea similar result also presents on the conditions offinishing gauge. Fig. 4 shows the morphology comparison of the sample(45 carbon steel) surface machinedin kerosene and air on finishing condition( pulse dura-tion 10 Hs, peak current 16 A, compressive air pres-sure is 0. 4 MPa). As shown in Fig. 4 that thfew micro cracks on the workpiece surface in(a)In keroseneof conventional EDM, while. it is not founcracks on the specimen surface processed in air, andwhich illustrates that smaller stress is produced in dryEDM(b)In airFig, 3 SEM photos of the sample cross-sectionTab. 3 The experimental conditions of the samples for(a)In keroseneSEM analysisvaluePulse duration/μsPulse interval/HsServo reference voltage/VIn dry EDM, discharge passages expand veryquickly because of the good expansibility of gas, whicheads to lower energy density in passages comparewith that of in liquid dielectric, where discharge pages are restricted by liquid be of lesser expansibility. Fig中国煤化工This means the workpiece surface machined by conven-CNMHGgetional EDM method achieve higher heat quality on unitrea. Furthermore, the therThe above phenomena can illustrate that small257Journal of Harbin Institute of Technology(New Series), VoL 16, No. 2, 2009tension are produced in dry EDM, and it may be con- diagram of wear testing is shown in Fig. 6. The materijectured that the surface obtained by dry EDM method al of the abrasion wheel is GCrl5 and its diameter ismay own better fatigue resistance performance than that 40mm with the speed of 200 rpm. The contact com-of conventional EDMthis point of view pression force between testing specimen and the abraneeds tested furthersion wheel is 40N, and the friction between them is dry2. 4 Surface hardnessfriction. The testing time for wearing samples is sameSurface micro hardness testing experiments were and after test experiment, the grinding cracks on theconducted for the samples EDMed in kerosene and air samples surface were measured and they are used as e-compared with that of the matrix that were processed by valuating specimens surface wear resistance perform-cutting. The EDMed samples were processed on theancesame machining parameters which is similar with theparameters shown in Tab. 3 except that servo referencevoltage is 100 V, and the surface hardness testing re-sults are shown in Fig. 5Fig 6 Schematic diagram of wear testingIn dry EDM, copper and titanium alloy (tC4Fig. 5 Surface micro hardness machined in different meditool electrode were used, and in conventional EDMcopper used as tool electrode. Samples were 45 carbonsteel and the size of the specimen was:(6 mm xFig. 5 shows that the surface hardness value 6 mm x 20 mm. The specimens were processed usingEDMed in kerosene is the greatest, and the hardness conventional EDM and dry EDM and the machining pavalue of the surface processed by dry EDM is seconda- rameters were: pulse duration 150 us, pulse intervalry, and the hardness value of matrix is the smallest. 64 us, peak current 20 A, and servo reference voltageThe similar hardness test results of dry EDM and con- 90 V. In dry EDM, the air pressure is 0. 5 MPaventional EDM on finish machining conditions haveig. 7 shows the specimensphotos after wearingsame rule as the results of rough machiningtests. It is found that the specimen processed in airThe experimental phenomena are explained aswith TC4 tool held the smallest grinding crack, and theWe could consider the cool velocity of the workpiece sample machined in kerosene with copper tool heldare fast whatever it is processed in kerosene or in air, second grinding crack, next, was the sample processednd the cooling course of the workpiece is similar to thquenching effect, which can increases the hardness ofthe workpiece surface. Because theance of air is weaker than that of kerosene, cooling ve-locity of the melt metal on discharge points in dry EDMis slower than that of in conventional EDM. then thealue of dry EDmthan that of convention EDM. If workpiece are proCoppercessed in kerosene, a certain quantities of C elementre generated on workpiece surface, such as Fe, CFig 7 Samples after wear teswhich induces higher hardness value of conventionalEDM compin air with copper tool, and the grinding crack width ofEDMth the surface processed by ary the matrix sample was the biggest. the test results in-2.5 Surface Wear Resistance Performancedicate workpiece machinedir with titanium allotThe surface wear resistance performance experi- poss中国煤化工 tance performancements were carried out among the specimens those were and-est wear resistancemachined with different tool electrode different work-CNMHGing dielectric and matrix. The wear tests were per-If titanium alloy used as tool, the heat quantityformed on wear testing machine tool and the schematicproduced by discharge difficultly conduct out of tool258·Journal of Harbin Institute of Technology(New Series), vol 16, No 2, 2009surface duelow thermo-conductivity of titani- and thinner recast layer, this advantage is helpful inum, whicheasily innduce thermal electron emission of appliccationthe titanium tool if pulse interval is not long, this4)Surface micro hardness value of dry EDM wasmeans that the heat erosion easily takes place on titani- greater than that of the matrix and was smaller than thatum alloy, and then titanium can depositof conventional EDMTherefore, a thin alloyingof titanium isWear testing experiments indicate that the sur-formed on the workpiece surface, so the surface pos-face wear resistance performance is related with the toolsesses better wear resistance performance. Fig. 8 material and the dielectric. In this study, surface proshows the surface energy spectrum analysis result of the cessed using dry edm with titanium alloy possessed thesample machined in air with titanium alloy tool, and it best wear resistance performance, and secondary wasdepicts titanium element existing on the surface (O el- the surface processed in kerosene, then was the surfaceement is also detected). Anyway the specimen pro- processed in air, and last was matrixcessed with titanium alloy tool holds better wear resistance performance. The surface processed in keroseneReferenceswith copper tool owns secondary surface wear resistance [1] Kunieda M, Furuoya S. Improvement of EDM efficiencyperformance. It is thought that higher hardness valueby supplying oxygen gas into gap. CIRP Annals, 1991, 40ormationof Fe, C contribute belter surface wear re1):215-218sistanceperformanceoftraditionalEdmthanthatofL2]LiLiqing,GuoYongfeng,baiJicheng,etal.compArativesample EDMed in air with copper electrode. There-experimental study of machining characteristics of air-mixedEDM and conventional EDM, Journal of Harbin Institute offore, the experiment illustrates the surface wear resist-Technology( New Series), 2007, 14(2): 170-173ance has relation with the tool material and the work [3] Masanoni Kunieda, Masahiro Yoshida. Electrical dischargeelectricachining in Gas. Annals of the CIRP, 1997, 46(1): 143[5] Kansal H K, Singh S, Kumar P. Technology and7.7elopments in powder mixed electric discharge maPMEDM). Journal of Materials Processing Te2007,184(4):32-41[6 Yan Biing Hwa, Tsai Hsien Chung, Huang Fuang Yuan.The effect in EDM of a dielectric of a urea solution in wateron modifying the surface of titanium. Intermational Joumalof Machine Tools Manufacture, 2005, 45(2): 194180200Fig8 Energy spectrum analysis of TC4 electrode[7 Miller S F. Shih A J, Qu Jun. Investigation of the sparkcycle on material removal rate in wire electrical discharge3 Conclusionsmachining of advanced materials. Intemational Journal ofMachine Tools and Manufacture, 2004, 44(4): 391-400e 1)The surface machined in gas has no metal lus- [8] Casas B, Torres Y, Llanes L. Fracture and fatigue behav-If the workpiece is anode, its surface appearanceior of electrical-discharge machined cemented carbides. In-temational Joumal of Refractory Metals and Hard Matershows black after processed in air or oxygen gas, andals,2006,24(1):l62-167ome debris are easily adhibited on workpie[9] Tao J, Shih A J. Dry and near-dry electrical dischargeit is necessary to strengthen debris expel in dry EDMmilling process. The 15th Intermational Symposium on2)Workpiece surface machined in air contain aElectromaching. Pittsburgh. 2007. 281-286.certain quantity 0 element. X-ray diffraction analysis 10 JEkmekci B, Tekkaya A E, Erden A. A semi-empirical ap-indicates FeO was generated on the surface machinedch for residual stresses in electric discharge machining( EDM). International Journal of Machine tools and Manu-In alr3)Compared with the surface processed in ker-facture,2006,46(6):858-868osene. the surface machined in air held lesser cracks中国煤化工CNMHG