外文翻译原文.pdf

Design, Manufacturing, andCalibration Process of One PieceLathe Dynamometer for Measurementin Two AxesErgun AtesC24Faculty of Engineering and Architecture,Department of Mechanical Engineering,The University of Balkesir,Balkesir 10145, Turkeye-mail: .trKadir AztekinInstitute of Sciences,Division of Mechanical Engineering,The University of Balkesir,Balkesir 10145, Turkeye-mail: .trMany pieces in structure of a dynamometer may negatively affectthe accuracy of the measurements since each piece may display adifferent strain and stress depending on its material and construc-tion. A one piece dynamometer has been designed to eliminatethese negative effects and to facilitate reliable force measure-ments. A dynamometer that its dimensions were confirmed follow-ing strength calculations was manufactured and calibrated in twodifferent ways. The first one was multipoint calibration method inwhich certain loads were applied to the dynamometer and strainvalues corresponding to these loads were matched. The secondcalibration method was implemented using Kistler dynamometerthat its results are accepted to be accurate by everyone and it wasbased on equivalency of force values resulted from work pieceprocessing and force values resulting from machining work pieceswith the same parameters to the manufactured dynamometer. Themanufactured dynamometer was capable of measuring cuttingforces and feeding forces. DOI: 10.1115/1.4024532Keywords: dynamometer, measurement, machinability, CAD-CAM1 IntroductionResearches have been focused on cutting forces since theseforces directly influence generation of heat, tool wear, power con-sumption, surface quality, and geometric tolerance of the processedwork piece 1,2. Many theoretic models were developed to mea-sure cutting forces. These models do not provide reliable resultssince machining has many complex parameters. Therefore, meas-uring cutting forces by experimental methods in unavoidable. Thereare many research works regarding measurement of cutting forcesin literature and many various dynamometers have been developedfor various machining methods. In a work in which the effects ofcutting tool rake angle on cutting forces in turning process wasstudied, it was indicated that cutting forces reduced as rake angleincreased. Cutting forces were measured by a dynamometer thatwas manufactured using two beam type load cells 3. It wasnoticed that a dynamometer, which could be used in turning andmetal cutting, manufactured in shape of an octagonal ring was ca-pable of measuring static and dynamic cutting forces accurately4. A particular type lathe tool dynamometer, which could be usedin ultra sensitive machining to measure statical and dynamicalforces was designed, manufactured and tested 5. A piezo-electriccell based dynamometer capable of measuring forces in three axisduring turning process was installed to a computer numericallycontrolled (CNC) counter for this work. The software was capableof analyzing, confirming, observing and recording data in real time6. A strain gauge base power dynamometer was designed andmanufactured for this work. Alternative tool method was producedby developing a recovery method with computer support using bal-ances that were produced to display the relationship between cut-ting forces and dimension errors 7. It was observed that majorityof these models use ring shape octagonal spring elements withstrain gauges attached to them. Negative effects of connection ele-ments used in attaching construction elements together whenassembling tool holder type dynamometers are substantial withrespect to sensitivity of the dynamometers. A one piece turning dy-namometer with strain gauge capable of measuring cutting forcesin two dimensions (axis) was designed and manufactured in thiswork and its calibration tests were implemented.2 Materials and Methods2.1 Design and Strength Calculations of the Dynamometer.The dynamometer was designed (DM) as a one piece material(Fig. 1(a)8,9. Solid-Works program was used for design works.Dimensions of the tool holder section that held the cutting tool,which was used, were planned as 25mmC225mm. Therefore, flexi-bility to bring the tools edge into same level with the mobile tail-stocks edge was provided by designing the section on which thetool holder will be located with 31mmC231mm dimensions.Standard length of the tool holder passes through the neck locatedinside the dynamometer and it establishes a tight connection withdynamometers body. The two perpendicular internal surfaces ofthe tool holder are completely full and void free. The other twosurfaces were connected to the dynamometers body using a totalof ten screws. Flow strength of AISI 1020, which was accepted asdynamometer material in the equation, was considered with doublesafety. The maximum load value to be applied to the system wascalculated to be 12883.904N based on the safety strength. For cut-ting depth of 2.5mm in a work safe bending strength was calculatedto be 19.2MPa in section strength control considering that the valueof the forces to be encountered were estimated to be 2000N. Stressand strain were calculated to be 0.234MPa and 1.112C210C06for100N. When these strain values were taken into consideration thedynamometer design used here was concluded to be compatiblewith necessary criteria for measuring the smallest strain values. Thedeflection in the edge point was calculated as follows when the toolholder was loaded at the edge point of a bar with square cross sec-tion 1,2. Maximum load was taken into consideration hereby anda deflection valued at 0.001776mm may result in the edge point.This value was a lot less than acceptable deflection values that mustoccur at the edge of the tool. Spring constant was calculated as9241.573 MN and it was placed in Eq. (3) and the value of theFig. 1 Designed (a) and manufactured (b) dynamometerContributed by the Manufacturing Engineering Division of ASME for publicationin the JOURNAL OF MANUFACTURING SCIENCE AND ENGINEERING. Manuscript receivedApril 6, 2012; final manuscript received May 4, 2013; published online July 17,2013. Assoc. Editor: Suhas Joshi.Journal of Manufacturing Science and Engineering AUGUST 2013, Vol. 135 / 044501-1CopyrightVC2013 by ASMEDownloaded From: / on 03/07/2014 Terms of Use: /termsnatural frequency was calculated as 784.482Hz. For example,when the largest value of revolution was 1000 rev/min (rpm) andthe frequency was 16.667 Hz the natural frequency of the dyna-mometer shall approximately be at least 21 times larger than thenormal working frequency. Thus, the criteria in which the naturalfrequency of the dynamometer must be at least 4 times less than itsnormal working frequency was fulfilled greatly.2.2 Manufacturing of the Dynamometer, WheatstoneBridge Circuit, and Data Collection System. The dynamometerwas manufactured using lathe, CNC milling machine, electro ero-sion machine and columnar drill workbench (Fig. 1(b). Manufac-turing steps of the dynamometer in order: Turning front surfaces ofcylindrical raw material in lathe, changing dimensions of the cylin-drical raw material into the desired dimensions in lathe, machiningthe lower section to which the tool holder will be connected usinglathe. CNC milling machine processes: Machining the outer surfa-ces of the section to which tool holder will be connected in CNCmilling machine, machining the lower section of the dynamometerin the CNC milling machine. Drill workbench process: The interiorof the quadrangle section was removed by drill as a process priorto square cross sectioning. Electro erosion machine processes: Theinterior surface of the section to which the tool holder would beconnected was removed in square cross section throughout itslength by electro erosion. Drill workbench processes: Openingscrew holes on two surfaces on the rectangular section by drill andtapping in order to fix the toll holder properly.A total of four strain gauges including one under and one on topof the dynamometer to measure the cutting forces and one at rightand one at left to measure the feeding force were fixed using an ad-hesive with appropriate components. The specifications of thestrain gauge were as follows: type FLA-10-11, gauge factor2.096%1 and its resistance was 12060.3 X. In order to connectthe strain gauges in the form of a half Wheatstone bridge circuit aparticular data cable was used. Specifications of the data cablewere as follows: 1206%0.1 resistance, trimmer potentiometerwith 15 turns and 50 X. The connection terminals located on thecable were appropriate for half or quarter bridge connections.Strain zeroing adjustment trimmer potentiometer, bridge comple-tion resistances and screwed terminals were inside the cable. Aclear container made of Plexiglas material with 4mm width wasdesigned to keep strain gauges from getting damaged by metalchips and external effects. The dynamometer with strain gaugesattached, completed solder connections and connected container,tool holder and tool are shown in Fig. 2 8,9. The specifications ofthe data collection system were: measurement resolution 16bit(610V/65,536 steps/8 channels), data collection speed eight sam-ples/second, channel gain may be adjusted between 1 and 890, datarecording in MS Excel, dynamic graphic axis adjustment, multi-channel zeroing, establishing mathematical processes between thechannels and multipoint calibration. One channel was arranged forcutting forces and another was arranged for feeding forces. Thedata collection system was connected to the computer using a USBcable. Specifications of this device connector were USB 2.0 andRS485 interface and it was controlled with a microprocessor. Soft-ware that was supplied with the data collection system was in-stalled on the computer. This facilitated real time observation andrecording of the data transferred from the dynamometer.2.3 Calibration Apparatus for Statical Loading Systems. Acalibration apparatus was designed and manufactured (Fig. 3(a).In order to calibrate the dynamometer with regard to cuttingforces it was fixed to horizontal table (Fig. 3(b) and it was fixedon perpendicular plate while being turned for 90deg for calibra-tion with respect to feeding forces (Fig. 3(c). Cutting force of thedynamometer was initially calibrated in the statical loading sys-tem. Multipoint calibration method was used in the software.Weight of the experimental mechanism was arranged to be 100Nand loading had started from 0N and it had continued untilaccomplishing 1000N in intervals of 100N. Microstrains corre-sponding to each 100N were recorded. The loads were reduced inFig. 2 Cabled dynamometer with identified strain gauge thatits strain gauge safeguard, tool holder, and tool are attached (inthe figure: 1- dynamometer, 2- strain gauge, 3- cable, 4- con-tainer, 5- tool holder, and 6- tool)Fig. 3 Calibration apparatus (a), cutting force calibration (b), and feeding force calibration (c) (in the figure: 1- feedingforce calibration table, 2- cutting force calibration table, 3- load frame, 4- load heading, 5- tool holder, and6- dynamometer)044501-2 / Vol. 135, AUGUST 2013 Transactions of the ASMEDownloaded From: / on 03/07/2014 Terms of Use: /termsintervals of 100N to accomplish 0N once again after 1000N wasaccomplished. Microstrains corresponding to each 100N reduc-tion were recorded. The waiting time for each loading was 30s.The experiments were repeated for 3 times for each loading andtheir average values were taken into consideration. The sameprocesses were repeated for calibration of the feeding force.2.4 Metal Chip Removing Processes. The dynamometer,which was designed and manufactured, was defined as DM(Fig. 4(a). Results of machining manufacturing (external surfaceturning) implemented using DM dynamometer were compared tothe machining results using Kistler-9257B dynamometer (Fig.4(b). The experiment set included Kistler-9257B dynamometer,Kistler Multichannel Charge Amplifier-5070 and DynoWare-DataAcquisition. The bench used for machining processes was theUniversal lathe Bench. The tool holder and tool displayed inFig. 2 were used for processes using both dynamometers. Thework bench used for the two dynamometers was the same basedon cutting speed, revolution number, cutting depth and machininglength parameters and it wasnt changed. The only parameter thathad changed was feeding Machining parameters are provided inTable 1.3 Results and Discussion3.1 Statical Calibration of DM Dynamometer. Forces cor-responding to cutting and feeding related to DM dynamometer areprovided in Table 2, the average strain calibration results are pro-vided. Maximum strain values in loading data were 6.35 le forcutting and 9.73le for feeding. R2 value of the certainty coeffi-cient was used in explaining the variety observed in the data as ascale for efficiency of the regression equation. R2 value was cal-culated as 0.9996 for cutting force and it was calculated as 0.9979for feeding force in this experiment 10. Regression lines, regres-sion equations and R2 values for cutting and feeding forces areprovided in Fig. 5. The R2 values were rather close to ideal valuesand this indicates that the relationship between the data was posi-tive and strong. The dynamometer has displayed the expectedbehavior in the process and it was observed that it could be a reli-able tool to measure cutting and feeding forces after completionof calibration process.3.2 Work Piece Machining for Calibration of Kistler andDM Dynamometer. Machining values used in this experimentare provided in Table 1. Diameter of AISI 1050 work piece wasreduced to 49mm from 50mm by turning its outer surface prior tothe experiment. A cutting tool with width of 5mm was used onthe work piece with diameter of 49mm channels were created inintervals of 7mm to reduce the diameter to 42mm. The dyna-mometers and their data collection systems were different,although all other parameters were the same during the machiningprocess. The cutting (Fc) and feeding (Fv) forces measured byKistler and DM dynamometers as a result of machining the workpiece are provided in Table 3.Fig. 4 Machining using (DM), the manufactured dynamometer (a) and Kistler-9257B (b)Table 1 Parameters used in machining with DM and Kistler dynamometersRPM (rev./min.)1000Cutting speed (m/min.)146.143Rake angle (deg)6 degInclination angle (deg)0 degApproach angle (deg)90 degDiameter (mm)49Last diameter (mm)44Tool holderCTGPR-2525M16Tool and coatingTPGN-160308-M30Unit specific cutting force1750N/mm2Cut of depth (mm)2.5Work materialAISI 1050Tool length (mm)150Tool coating materialsTiCN-Al2O3-TiNMaterial coefficient0.27Feed (mm/rev.) 0.08 0.12 0.24Table 2 Statical calibration values of DM dynamometerForce (N) 0 100 200 300 400 500 600 700 800 900 1000Cutting (le) 0.0000 0.6523 1.3639 1.8976 2.5499 3.2023 3.8546 4.5069 5.0999 5.6336 6.3452Feed (le) 0.0000 1.0679 2.0764 3.1443 4.3901 5.1613 6.1698 7.1190 8.0094 8.9581 9.7293Journal of Manufacturing Science and Engineering AUGUST 2013, Vol. 135 / 044501-3Downloaded From: / on 03/07/2014 Terms of Use: /termsFc and Fv values were indicated as microstrains for DM dyna-mometer in order to calibrate DM dynamometer based on Kistlerdynamometer. The strain values that resulted from DM dynamom-eter under load effects were recorded along with the force values(Fc and Fv) produced by Kistler into a calibration file using multi-point calibration method by the software. It was concluded thatthe feeding and cutting forces obtained by DM dynamometerwere exact matches for cutting and feeding forces measured byKistler. Behaviors of the dynamometers under load effect weredifferent. Values obtained by DM dynamometer were changedinto force values using Kistler dynamometer and the calibrationwas implemented. When DM dynamometer was used with thesaid calibration file the same strain values were obtained underthe same machining conditions and therefore, the force values cor-responding to these values were also the same with the force val-ues measured by Kistler dynamometer. Fv values that weremeasured in this experiment recoded a change depending on raisein values of the feeding. All feeding forces were smaller than cut-ting forces. In order to determine the ratio between the cutting andfeeding forces values calculated based on Fc/Fv were provided inTable 3. When Fc/Fv ratios were studied it was noticed that theseratios increased as feeding during machining with Kistler and DMdynamometers increased. The smallest and largest ratios for Kis-tler were calculated to be 1.547 and 2.150 and these values were1.900 and 2.429 for DM dynamometer. Theoretical cutting forceswere calculated using the machining parameters indicated inTable 1 and these values were indicated in Table 3 as comparisonsto force values measured by Kistler dynamometer. Statically talk-ing 5% error was an acceptable error rate for engineering calcula-tions 10. Should the theoretical cutting forces be considered asreference values and should the % differences between the valuesbe studied, the largest would be C03.75% and the smallest wouldbe C01.76%. Table 4 shows that there were differences betweenvalues obtained for cutting forces calculated based on theoreti-cally accepted values and the experimental values. Therefore,Fig. 5 Variation of cutting and feeding forces with microstrain for DM dynamometer calibrationTable 3 Force and strain values accomplished with Kistler andDM dynamometers during machiningKistler dynamometer(force)DM dynamometer(microstrain)Feed(mm/dev.) Fc (N) Fv (N) Fc/Fv Fc (le)Fv(le) Fc/Fv0.08 672.99 434.91 1.547 2.556 1.345 1.9000.12 869.14 507.81 1.712 3.268 1.672 1.9550.24 1404.09 652.95 2.150 7.132 2.936 2.429Table 4 Comparison of the theoretical and experimental cut-ting forcesFc