3y ago
503.10 KB
8 Pages
Last View : Today
Last Download : 5m ago
Upload by : Wade Mabry

ABRASIVE JET MACHINING FOR EDGE GENERATIONMatthew W. Chastagner and Albert J. ShihMechanical EngineeringUniversity of MichiganAnn Arbor, MichiganKEYWORDSstudy demonstrates that a B-spline can providea better fit and geometrical description of theedge profile.Abrasive Jet Machining, Edge Finishing, EdgeGeneration, Erosion WearINTRODUCTIONABSTRACTAbrasive jet machining (AJM), also calledabrasive micro blasting, is a manufacturingprocess that utilizes a high-pressure air streamcarrying small particles to impinge the workpiecesurface for material removal and shapegeneration. The removal occurs due to theerosive action of the particles striking theworkpiece surface. AJM has limited materialremoval capability and is typically used as afinishing process (Gillespie 1999).The edge with a consistent and precise shapeis important for highly stressed mechanicalcomponents. This study investigates thegeneration, measurement, and definition ofedges. Abrasive jet machining, a flexibleprocess ideal for difficult-to-reach areas, isapplied for edge generation. A conoscopy laserwith small, 25 μm spot size is scanned acrossthe edge for measurement. A B-spline curve isapplied to fit the edge profile by an optimizationmethod. Silicon carbide media, 50 μm averagesize, was used to erode a 90º edge on aworkpiece of Inconel 718. Effects of blastingtime, stand off distance between edge andnozzle, and orientation of a nozzle are studied. Itis found that the edge radius is limited to below0.15 mm using abrasive jet machining. Underlong blasting durations, the edge radius does notchange, but collateral damage around the edgeis significant. Long standoff distances and highangles of blasting are beneficial in reducing thelevel of collateral damage. Edge radius iscommonly used as the only parameter toquantitatively describe the edge profile. ThisTransactions of NAMRI/SMEAJM is advantageous in two aspects. First, ithas a high degree of flexibility. The abrasivemedia can be carried by a flexible hose to reachinternal, difficult-to-reach regions. Second, AJMhas localized force and less heat generationthan traditional machining processes. In thisstudy, AJM is investigated to generate a desirededge shape.In highly stressed mechanical components,such as the turbine blades and rotors in aircraftengines it is important to avoid sharp edges,which can lead to cracks and premature partfailure due to localized stress concentrations.359Volume 35, 2007

The technical challenges are the definition,measurement, and generation of edges with thedesired shape. In difficult-to-reach edges, forexample, at the intersection of holes inside thepart, the generation and measurement of theedges are particularly challenging due to thegeometrical constraints for limited toolingaccess. AJM, abrasive flow machining and turboabrasive machining are all suitable for internaledge generation because of their ability to entersmall holes and reach the internal edges(Gillespie 1999). The investigation into the useof AJM was undertaken since it can be applieddirectly to a single controlled location.measured edge profile. The level of fitness isquantified using an error index.In this paper, the experimental setup of theAJM machine, blasting media, samplepreparation, and laser edge measurement arefirst introduced. Effects of three key processparameters, the time of blasting, the stand offdistance, and the angle of impingement, on theedge radius and collateral damage around thecreated edge are examined. Finally, themathematical analysis of edge profile is studied.A variety of applications of AJM have beenstudied. Balasubramaniam et al. (1999, 2000)investigated using AJM to remove burrs at theintersection of cross-drilled holes on stainlesssteel, while also examining the subsequent edgeradius. Lemaster et al. (2005) and Qu et al.(2005) have studied the use of AJM as a surfacefinishing process on wood and WC-Cocomposite, respectively. Others have examinedthe AJM process for glass etching and groovingin micro-systems parts (Park et al. 2004) and flatpanel displays (Slikkerveer et al. 2000). Canby(2003) has reported AJM for removing burrs onvery small holes in aircraft turbine blades.Balasubramaniam et al. (2000) has proposed amathematical relationship for edge radiusdefinition when using AJM on a blunt surface.However, this method does not take intoaccount the edge radius created on a predefinedsharp edge. Therefore, the investigation into thedefinition of an AJM generated edge hasbecome one of the goals of this research.Another goal of this study is to develop theopticalnon-contactconoscopylasermeasurement with a small, 25 μm, focal pointspot size to quantify the shape of the edge afterAJM under different processing conditions.The AJM setup, edge measurement,experimental procedure, and edge profileanalysis are discussed in the following sections.EXPERIMENTAL SETUPAJM Machine SetupThe AJM machine is a Comco micro blaster,Model MB1000, in combination with a collector(Comco WS2200-1) for removing the spentmedia. This machine has an oscillatingmechanical valve to meter the media into the airstream (Weightman 1970). This method is morecomplicated than using a venturi to draw theparticles into a mixing chamber (Lemaster et al.2005). The Comco machine design has theadvantage of metering a small and preciseamount of micro-size media for AJM. Themixture of media and airflow is then directed to anozzle where it exits towards the part, as shownin Figure 1(A).This AJM machine has an internal timer,allowing for the duration of the blast to becontrolled precisely. A special fixture wasdesigned and built to orient the nozzle andworkpiece, which is defined by two parameters:the distance from the edge to the center of thenozzle and the angle between the nozzle andthe edge, denoted as l and α in Figure 1(A),respectively. This fixture consisted of atoolmakers vise with an attachment for holdingthe nozzle perpendicular to the vise, as shown inFigure 1(B). Special machined jaw faces weremade for the vice to hold the sample at thecorrect position. For tests in which the nozzleangle relative to the sample needed to bechanged, angled inserts were used to orient theworkpiece without moving the nozzle.In the cutting tool industry, the shape of theedge is important for tool performance. Theradius and waterfall shape are two commonlyused features to define an edge (Shaffer 1999).However, the mathematical description of anedge has not been studied extensively. Usinghigh resolution edge measurement, such as theconoscopy laser used in this study, can give amore detailed quantitative analysis of the edge.There is a need for a more precise definition ofedge than just a radius. In this study, amathematical model based on B-spline isinvestigated to provide a better fit to theTransactions of NAMRI/SME360Volume 35, 2007

(Brother HS-5100) was used to cut the Inconel718 workpiece to create the sharp edge.Dimensions of the sample, as shown in Figure1(B), were 6.40 mm x 6.40 mm x 20.0 mm. AfterAJM, the edge is eroded, to produce a roundedtip and collateral damage is present on bothsides of the edge. As shown in Figure 1(C), theedge radius is marked by r, the width and depthof the collateral damage are denoted as w1 andc1 on the left side (close to the nozzle) and w2and c2 on the right side (away from the nozzle).These parameters are applied to quantify theedge after AJM.FIGURE 2. SEM MICROGRAPHS OF SiC MEDIA.Edge MeasurementsFigure 3(A) shows the setup to measure theedge profile using a conoscopy laser sensor(Optimet Smart ConoProbe #25). This sensorwas chosen because it: (1) has a small, 25 ȝmfocal point, (2) is precise with sub-μm accuracy,and (3) allows for a high incidence angle, up to80º, in measurement. A computer-controlledstage, made by Aerotech , is applied to movethe sensor relative to the workpiece.FIGURE 1. (A) CONFIGURATION OF THE SETUPAND KEY PARAMETERS, (B) PICTURE OF THENOZZLE AND WORKPIECE, AND (C) WORKPIECEEDGE AFTER BLASTING WITH KEYGEOMETRICAL PARAMETERS.The nozzle is made of tungsten carbide andhas a round inner diameter, marked as d inFigure 1(A). In this study, d is 1.5 mm. Siliconcarbide (SiC) media with a 50 ȝm average sizeis used as the blasting media. According to theguideline of the AJM machine builder, the nozzleinner diameter needs to be 20 times larger thanthe average media size. The SiC was selecteddue to its high hardness (2700 Knoop or 9 Mohr hardness), in combination with its sharpblocky shape. This allows for effective erosionwear of the Inconel workpiece. Figure 2 showsthe scanning electron microscope (SEM) (PhilipsXL30) micrographs of the shape and the size ofthe SiC media.SensorWorkpieceWorkpiece(A)(B)FIGURE 3. SETUP FOR THE EDGE PROFILELASER MEASUREMENT.The close-up view of the workpiece andsensor is shown in Figure 3(B). A line scan, withThe angle of the edge investigated in thisstudy is 90º. A wire electrical discharge machineTransactions of NAMRI/SMESensor361Volume 35, 2007

region to L1 and L2 are searched to find c1 andc2, respectively.RESULTSa scanning speed of 100 mm/min and 3000 datapoints per second sampling frequency, resultedin about 8000 data points per edge. This linescan was applied across the edge.SEM Micrographs of Edges After AJMExperimental ProcedureSEM micrographs of an AJM machined edgein Exp. I with t 15 s are shown in Figure 4(A).A close-up view of the edge and the depth ofcollateral damage are shown in Figure 4(B). Thenumber 1 is placed on the side with w1 and c1,i.e., the side close to the nozzle. The dashedline marks the edge profile measurement traceusing the conoscopy laser. The collateraldamage is obvious. Quantitative values ofcollateral damage will be discussed in thefollowing section (Exp. I). Figure 4(C) shows theless obvious collateral damage in the Exp. II testwith l 5 mm, t 7 s, and α 0º.This research studies the effect of three keyprocess parameters: l, α, and the time ofblasting, t. The input air pressure to the nozzlewas set at 552 kPa. Three sets of experiments,marked as Exps. I, II, and III, were performed. Exp. I maintained l 0 mm and α 0º, whilet was varied from 0 to 15 in 1 s increments.Another test was conducted under the sameconditions with t 30 s. This experimentstudies the effect of time of blasting.Exp. II investigates the effect of distancebetween nozzle and edge tip. The value of lwas varied from 0 to 5 in 1 mm steps, while t 7 s and α 0º.Exp. III studies the effect of angle α, whichwas varied from 0º to 30º in 5º increments(with l 0 mm and t 7 s).12Three repeated tests were conducted for everytest condition. All tests eroded a spot on theedge, as illustrated by spots on the workpieceedge in Figure 3(B). In production, the nozzle isexpected to move along the edge to generatethe desired edge radius. The spot test in thisstudy is the first step in understanding themagnitude of the edge radius and collateraldamage in AJM.(A)(C)Edge Profile AnalysisFIGURE 4. SEM MICROGRAPHS OF THE AJMREGION AND EDGES (A) EXP. I, 15 S, (B) CLOSEUP VIEW OF (A), AND (C) EXP. II WITH 5 MMSTANDOFF DISTANCE.The measured edge profile is recorded by thex-y cloud of data points and needs to beanalyzed to determine the edge radius and levelof collateral damage. To determine the edgeradius, two boundary points are first identified todefine the range for curve fitting an arc. A leastsquares curve fitting method was utilized to fit anarc to all of the data points between twoboundary points, thus finding the edge radius.Exp. I. Effect of Blasting TimeFigure 5 shows results of r, c1, c2, w1 and w2for the 16 tests in Exp. I. The error bars indicatethe deviation of the three repeated tests from theaverage value. The edge radius of 0.022 mm at t 0 represents the initial edge radius after wireEDM. The edge radius increases rapidly in thebeginning of blasting to 0.07, 0.08, 0.09, and0.10 mm after 1, 2, 3, and 4 s of blasting,respectively. As the edge takes its shape after 4s, the edge radius increases very slowly from0.12 to 0.13 mm. The long, 30 s blasting timeproduces a 0.14 mm edge radius, which is onlyTo determine the collateral damage, thestraight side lines unaffected by AJM, marked asL1 and L2 in Figure 1(C), are first identified.These two lines intersect as a point, A. Twoother control points are identified on the side lineto determine the distances w1 and w2,respectively, from point A. The maximumdistance from points in the collateral damagedTransactions of NAMRI/SME(B)362Volume 35, 2007

In conclusion, the Exp. I results in Figure 5show that most of the AJM edge generation istaking place within the first 4 s, and after this,extended blasting times only cause a gradualincrease of the depth and essentially no changein the width of the collateral damage.Exp. II: Effect of Standoff Distance0.2Figure 6 shows the results of r, c1, c2, w1 andw2 for the 6 tests in Exp. II with l varying from 0to 5 mm, while keeping t 7 s and Į 0 . Theedge radius remains about the same, 0.11 mm,independent of l. The collateral damage widthhas a slight increasing trend, but remains in the1.3 to 1.5 mm range. The depth of collateraldamage shows a steady trend of decreasing,from about 0.15 mm to 0.1 mm. The higheststand off distance (l 5 mm), with the ds to the SEM micrograph of the edgeshown in Figure 4(C)., r (mm)Depth of collateraldamage, c (mm) 16 20 24Blasting time (s)28FIGURE 5. EFFECTS OF BLASTING TIME ONEDGE RADIUS AND DEPTH AND WIDTH OFCOLLATERAL DAMAGE (EXP. I RESULTS).Depth of collateraldamage, c (mm)The width of collateral damage, w1 and w2,ranges between 1.2 to 1.4 mm after 2 s ofblasting. Blasting time does not change thewidth of collateral damage significantly after thefirst 2 s. Since α 0º, w1 and w2 shall be closeto each other. However, the difference betweenw1 and w2 indicates that an alignment errorexists. It is difficult to setup the exact α 0º andsuch discrepancy is 500Unlike the results of r, w1, and w2, the depth ofcollateral damage increases steadily and linearlywith respect to time. After 15 s of blasting, c1was 0.33 mm, while c2 was 0.26 mm. This lineartrend extends to the 30 s long blasting time,which produces collateral damage about 0.5 mmdeep. On average, each second of blastingerodes about 17 μm of material.Transactions of NAMRI/SME0.1032Width of collateraldamage, w (mm)Width of collateraldamage, w (mm)Radius, r (mm)slightly increased from the 15 s value. Thisindicates that the edge radius in AJM isdetermined in the initial stage. This immediatesteady state erosion has also been observed insteel (Tilly 1969). In Exps. II and III, 7 s hasbeen selected as the time of blasting because ofthe steady-state nature of the edge radiusgeneration.24Stand off Distance, l (mm)6FIGURE 6. EFFECTS OF NOZZLE STAND OFFDISTANCE ON EDGE RADIUS AND DEPTH ANDWIDTH OF COLLATERAL DAMAGE (EXP. IIRESULTS).Results in Exp. II point to a strategy of movingthe nozzle tip away from edge to create the363Volume 35, 2007

same level of edge radius while reducing thedepth of collateral damage.impingement is about 90 , i.e., the particlevelocity vector is perpendicular to the surface.On the contrary, for ductile materials, whichinclude the Inconel 718 used in this study, theshallow angle of impingement will result in highmaterial removal rate (Cousens and Hutchings1983; Williams 1994). As shown in Figure 1(A),the area away from the nozzle has a shallowangle of impingement and will result in morematerial removal.Exp. III: Effect of AngleFigure 7 shows the results of r, c1, c2, w1 andw2 for the 7 tests in Exp. III with Į varying from0 to 30 , while keeping t 7 s and l 0 mm.The edge radius remains within the 0.1 to 0.12mm range and does not vary significantly withrespect to the angle Į.The erosion process for ductile material canbe characterized by either a dentation-type,ploughing type or cutting type event (Morrison1986). The dentation-type event only serves tomove material, as it does not cut. Ploughing andcutting are the primary reason for the materialremoval. This removal occurs as the particlestrikes the material, causing small platelets ofmaterial to be extracted from the surface andexpelled. If the platelet is not fully removed,subsequent particles striking the surface willcause the platelet to finally become detached.This platelet removal process requires largeshear deformations coupled with heat increaseand very high strain rates (Shewmon 1981).Radius, r (mm)The nozzle orientation has a significant impacton w1 and w2. As the angle Į is stepped up from0 to 30 , on the side close to the nozzle, w1decreases and, on the side away from thenozzle, w2 increases. This is caused by theangle of impingement θ, as illustrated in Figure1(A), of the particles impinging the surface. of collateraldamage, c (mm)Width of collateraldamage, w (mm)04The angle Į also has a significant effect on thedepth of collateral damage. Both c1 and c2 showa trend of decreasing at high Į. This is apleasantly surprising observation. It can beexplained by the angle of impingement. In theregion close to the nozzle, the angle ofimpingement is close to 90 and the materialremoval rate is low because the work-material isductile. In the region away from the nozzle, theangle of impingement is shallow; it can increasethe width of collateral damage (w2) but does nothave enough speed and energy to createsignificant material removal.w2w132100.25c1c20. II and III results show that it is possibleto strategize the l and α to reduce collateraldamage in AJM for edge generation.00102030Angle, Į (deg)40ANALYSIS OF EDGE PROFILEFIGURE 7. EFFECTS OF NOZZLE ORIENTATIONANGLE ON EDGE RADIUS AND DEPTH ANDWIDTH OF COLLATERAL DAMAGE (EXP. IIIRESULTS).Various edge profiles are generated in AJM.Figure 8 illustrates the close-up view of sevenedge profiles generated in Exp. III. Thehorizontal axis is the stage traverse distanceand the vertical axis is the sensor measureddistance. The profile shown in Figure 8 has beentranslated to make the lowest point in the sensormeasurement equivalent to 0 in both the stagetraversed and sensor measured distances.Erosion wear is the mechanism for materialremoval in AJM. The angle of impingementaffects the material removal rate significantly.For brittle materials, the maximum materialremoval rate occurs when the angle ofTransactions of NAMRI/SME364Volume 35, 2007

NSensor MeasurementDistance (mm)0.6Sensor MeasurementDistance (mm)0.6Sensor MeasurementDistance (mm)The test profile of α 10º in Exp. III is selectedfor further analysis. Two control points areselected to define the curve section for analysis.This section, marked in Figure 8, is rotated andre-oriented, as shown in Figure 9. Two curvefitting methods, a least-squares fitting of acircular arc and B-spline fit, are utilized tomathematically describe this edge profile.0.6miThe B-spline curve has the freedom to selecttwo control points to fit the edge profile. The B spline is optimized using Matlab to find thelocations of control points, while minimizing theerror index e. The circular arc and B-spline inFigure 9 have e 1.6 and 4.4 μm, respectively.The B-spline has a better fit, which can also berecognized in Figure 9. One of the advantagesof using circular arc is the result of a simple,single value for the edge representation andcomparison. It remains as a technical challengeon how to better represent the B-spline curves.0.20.010ºBoundary pointsNwhere N is the number of sample pointsmpselected on the edge profile, xi and xi are themeasured and curve fit predicted points of edgeprofile, respectively.0.40.4 xip ) 2i 1e 5º0º (x15º0.20.00.1825º20ºSensor MeasurementDistance (mm) -0.2 0.00.2Sensor MeasurementDistance (mm)Stage TraverseDistance (mm)Measured edge profileArc fitB-spline fitControl points for

Abrasive jet machining (AJM), also called abrasive micro blasting, is a manufacturing process that utilizes a high-pressure air stream carrying small particles to impinge the workpiece surface for material removal and shape generation. The removal occurs due to the erosive action of the particles striking the workpiece surface. AJM has limited material removal capability and is typically used .

Related Documents:

In Abrasive Jet Machining (AJM), abrasive particles are made to impinge on the work material at a high velocity. The jet of abrasive particles is carried by carrier gas or air. High velocity stream of abrasive is generated by converting the pressure energy of the carrier gas or air to its kinetic energy and hence high velocity jet. Nozzle directs the abrasive jet in a controlled manner onto .

AWJM, the abrasive particles are allowed to entrain in water jet to form abrasive water jet with significant velocity of 800 m/s. Such high velocity abrasive jet can machine almost any material. Fig. 1 shows the photographic view of a commercial CNC water jet machining system along with close-up view of the cutting head.

In Abrasive Jet Machining (AJM), abrasive particles are made to impinge on the work material at a high velocity. The jet of abrasive particles is carried by carrier gas or air. The high velocity stream of abrasive is generated by converting the pressure energy of the carrier gas or air to its kinetic energy and hence high velocity jet. The nozzle directs the abrasive jet in a controlled manner .

The abrasive water jet machining process is characterized by large number of process parameters that determine efficiency, economy and quality of the whole process. Figure 2 demonstrates the factors influencing AWJ machining process. Shanmugam and Masood (2009) have made an investigation on the kerf taper angle, generated by Abrasive Water Jet (AWJ) machining of two kinds of composite .

Abrasive water jet machining (AWJM) process is one of the most recent developed non-traditional machining processes used for machining of composite materials. In AWJM process, machining of work piece material takes place when a high speed water jet mixed with abrasives impinges on it. This process is suitable for heat sensitive materials especially composites because it produces almost no heat .

process parameters, analysis of machining, response characteristics, Applications of the Abrasive Jet Machining, Water Jet Machining, Abrasive water Jet Machining, Ultra sonic machining. . Ghosh, Amitabh., Manufacturing Processes. New Delhi: Tata McGraw Hill

Eastern Oak Looper (Caterpillar) 27 ACE-jet, Azasol Gall midges ACE-jet Tussock moth (Caterpillar) 27 TREE-äge , ACE-jet, Azasol Anthracnose 23 PhOsPhO-jet, Alamo Aphids 23 ACE-jet, ImA-jet, or Azasol Bagworm 24 TREE-äge, ACE-jet, Azasol Japanese Beetle 31 ImA-jet, ACE-jet, Azasol Lace Bugs ACE-jet, ImA-jet Leaf spot Arbor-OTC

Abrasive jet Machining consists of 1. Gas propulsion system 2. Abrasive feeder 3. Machining Chamber 4. AJM Nozzle 5. Abrasives Gas Propulsion System Supplies clean and dry air. Air, Nitrogen and carbon dioxide to propel the abrasive particles. Gas may be supplied either from a compressor or a cylinder. In case of a compressor, air filter cum drier should be used to avoid water or oil .