- Email: [email protected]
Laser cleaning and dressing of vitrified grinding wheels
Journal of Materials Processing Technology 185 (2007) 17–23
Laser cleaning and dressing of vitrified grinding wheels M.J. Jackson a,∗ , A. Khangar b , X. Chen c , G.M. Robinson a , V.C. Venkatesh d , N.B. Dahotre b a
Center for Advanced Manufacturing, Purdue University, College of Technology, 401 North Grant Sreet, West Lafayette, IN 47907-2021, USA Department of Materials Science and Engineering, University of Tennessee-Knoxville, 326 Dougherty Hall, 1512 Middle Drive, Knoxville, TN 37996, USA c School of Mechanical, Materials, and Manufacturing Engineering and Management, University of Nottingham, University Park, Nottingham NG7 2RD, UK d Department of Mechanical Engineering, University of Technology Malaysia, 81310 UTM Skudai, Johor Durul Ta’zim, Malaysia b
Abstract Grinding is an industrial process that produces engineering components with a desired surface finish. Prior to the development of continuous dressing operations, the grinding efficiency of vitrified grinding wheels deteriorates as the sharp cutting edges become blunt due to the formation of wear flats. Dressing is essentially a sharpening operation designed to generate a specific topography on the working surface of the grinding wheel. The use of high power lasers is being explored as a non-contact cleaning and dressing technique. In the present study, a high power laser was used to clean metal chips from the surface of the grinding wheel and to dress the wheel by causing phase transformations to occur on the surface of vitrified grinding wheel. Experimental results indicated that laser modified grinding wheels are comparable in performance to conventionally cleaned and dressed grinding wheels. © 2006 Elsevier B.V. All rights reserved. Keywords: Vitrified grinding wheels; Laser materials processing; Grinding; Tool steels; Aerospace materials
1. Introduction 1.1. Conventional dressing of grinding wheels Grinding wheels are used for grinding materials with poor machinability because of their longer life, high grinding efficiency, and dimensional stability [1]. The grinding wheel also gets loaded with metal chips during machining [2]. The existent grinding tools wear at a very high rate and also generate high frictional heat at the abrasive grain-metal chip contact region, which leads to surface and sub-surface damage in the ground components. Also, there is a loss of form tolerance and dimensional stability as grinding progresses over a large number of grinding passes [3]. Ideally, the abrasive particles on the surface of the grinding wheel should get automatically sharpened when worn out, by either entirely detaching from the wheel face or by fracture, thus exposing new particles with sharper cutting edges. However, in practice both geometric and functional characteristics of the grinding wheel have to be restored periodically by dressing [2]. Dressing is a sharpening operation designed to generate a particular surface topography on the cutting face of
∗
Corresponding author. Tel.: +1 765 494 0365; fax: +1 765 494 6219. E-mail address: [email protected] (M.J. Jackson).
0924-0136/$ – see front matter © 2006 Elsevier B.V. All rights reserved. doi:10.1016/j.jmatprotec.2006.03.109
the wheel. Dressing of the worn out grinding wheel surface is done by various means to modify its surface topography, which in turn improves grinding efficiency [4]. Dressing significantly affects the quality of the ground product, as characterized by its size and shape, surface roughness and integrity [5]. The conventional contact-type methods, like mechanical dressing using a diamond dresser, results in excessive grinding wheel material loss. The bond fracture and abrasive grain break-off, as a result of crushing of the dresser are the material removal mechanisms for contact type processes. In fact, only 10% of the wheel by volume is removed during actual grinding, while the rest is removed during dressing operations [6]. Mechanical dressing, though effective, also induces stresses and causes deep cracks and undercuts. These factors eventually cause loosening of the chunks of grains and reduce the number of effective cutting edges [2]. The conventional processes do not produce consistent grinding results due to dresser wear which affects the wheel surface topography and its performance in grinding. A worn-out dresser cannot produce sufficient protrusion of cutting grain edges [1]. To obtain consistent grinding outputs, either the dresser geometry has to be maintained or suitable dressing conditions corresponding to modified dresser geometry have to be selected, which are impossible in contact type processes and also complicated because of dresser wear [4]. Yet in order to obtain
18
M.J. Jackson et al. / Journal of Materials Processing Technology 185 (2007) 17–23
consistent grinding results, a dressing procedure, which is reproducible in nature, is essential. To achieve these requirements, methods have been developed to maintain an optimized grinding wheel. These methods include: slurry and steel roll application, pressurized jet dressing using abrasive slurries, and air jet abrasive dressing. In air jet dressing, the dressing rate is controlled in three ways: (a) by controlling the pressure of the mixed air, (b) by controlling the ejecting nozzle traverse speed, and (c) by controlling the number of traverse passes across the grinding wheel. Therefore, the dressing rate will control the optimum grain protrusion height. 1.2. Laser cleaning and dressing of grinding wheels High power lasers that are currently used as a non-contact type machining tool for various manufacturing applications such as welding, drilling, cutting, etc., can also be used as a non-contact type dressing tool. The salient features of a laser include high intensity fluence, directionality, and spatial coherence, which can be used to process hard and brittle materials efficiently. Laser induced thermal processing leads to effects such as melting, vaporization, and plasma formation on the material of the grinding wheel, which can be exploited during the dressing procedure. When a laser beam irradiates the surface of a grinding wheel, it may be considered that the energy flows in one direction in a semi-infinite body. If there is no convection, or heat generation, the basic equation governing the flow of heat is, ∂2 T 1 ∂T = 2 ∂z α ∂t
(1)
Where T is the temperature in the grinding wheel, z the depth from the wheel surface, α the thermal diffusivity of the grinding wheel, and t is the time after laser irradiation commences. If it is assumed that the laser power flux input into the wheel is q, with no radiant heat loss or melting, then the solution to Eq. (2) is, √ 2q z √ Tz,t = (2) α t ierfc K 2 αt Where K is the thermal conductivity of the grinding wheel material, and ierfc is the integral of the complementary error function. Eq. (2) indicates that both q and t contribute to elevating the temperature of the surface of the grinding wheel. The depth of laser energy penetrated into the wheel surface is constrained by the duration of the laser pulse. Increasing irradiation time will allow the laser energy to penetrate deeper into the grinding wheel. For the purpose of laser dressing, a higher temperature is needed in order to remove metal chips and re-shape individual grinding grains. During laser dressing, the wheel surface topography of the grinding wheel is modified by melting of the material and subsequent re-solidification of a portion of the molten layer. During the process, rapid heating and cooling induces cracks in the re-solidified layer. The microcracks help remove the re-melted layer during grinding after a few initial grinding strokes, which then exposes new cutting edges. In laser dressing, the grinding wheel is subjected to high power laser intensity, which produces
craters on the surface and also induces microcracks in the re-cast molten layer. By using different laser powers during dressing, the extent of the re-cast molten layer can be controlled and controlling the dressing feed can vary the surface topography. Laser dressing removes the wheel material by ablation of the bonding material to expose sharp grains. When the heating time and energy density is selected correctly, the vitrified bond (glass phase) of the wheel is softened and even melted thus facilitating the removal of the bonding material [1]. Dressing of the grinding wheel by laser generates welldefined grooves and tracks on the surface of the wheel. Essentially, laser dressing produces microcutting edges due to the formation of microcracks on the worn-out grains. When these craters are formed on the bond, the grits are loosened and subsequently removed due to insufficient volume of load surrounding the grain [4]. Laser dressing thus has the advantage of being a non-contact type process in which selective removal of clogged material alone is possible by appropriate focusing of the laser beam on the selected portions of the wheel surface. Also the material wastage in terms of the debris produced during dressing operation, can be substantially reduced by the use of a laser as a dressing tool. Thus environmentally benign grinding operations can be made possible, which will help reduce the problem of disposing contaminated grinding debris. There are several inherent advantages associated with the use of laser for dressing applications. Laser dressing is a very fast process and it can be easily automated. Also, selective removal of the clogged material alone is possible and desired surface structure (roughness, grain morphology and porosity) can be generated. Furthermore, consistent dressing conditions can be produced by the use of laser and this can help achieve grinding reproducibility. As the laser beam can be delivered using a fiber optic cable, remote dressing operation without discontinuation of the grinding process during laser dressing is possible. Thus the downtime in the grinding operation associated with conventional methods, can either be eliminated or substantially reduced in laser dressing. Earlier studies [2,4,7] used pulsed laser to compare the laser dressing process with conventional mechanical dressing methods. The pulsed laser powers used were of the order of 1–5 × 1010 W/m2 and most of the work concentrated on comparison of the grinding performance of laser dressed wheels with that of diamond dressed ones. Laser-assisted simultaneous truing and dressing has also been attempted, to overcome the problems associated with mechanical dressing [8]. Though useful, the studies did not deal with the nature of physical changes taking place in the grinding wheel during interaction with laser energy, which happens to be a fundamental aspect to contribute towards its dressing performance. Prior to any dressing operation, the grinding wheel tends to load with metal chips that need to be removed. Wheel loading is one of the most common problems in grinding operations, particularly for grinding aerospace materials. As grinding continues, removed chips may adhere in the spaces between abrasive grains and deteriorate the cutting ability of the grinding wheel. A common method used to prevent the wheel from loading is achieved by delivering a large amount of coolant to the grinding
M.J. Jackson et al. / Journal of Materials Processing Technology 185 (2007) 17–23
zone. However, this consumes huge amounts of energy in coolant delivery, especially for high speed grinding processes. Maintaining and disposing of coolant is also an environmental issue and the costs involved are substantial. Another method, which is often used, is to remove loaded chip materials by dressing the wheel periodically, to restore a sharp wheel surface. However, dressing of grinding wheels with diamond not only causes excessive wheel loss but also interrupts grinding during dressing. In addition, the dresser wears away with time due to its direct contact with the wheel surface. Frequent use of dressing wheels is also not acceptable for super-abrasive wheels whose cost is considerably more than conventional grinding wheels. Lasers have been successfully applied to material removal processes such as laser cutting, and drilling. Laser cleaning techniques have been used in removing pollution layers from valuable artifacts and without damaging the delicate patina of the substrate [9]. This suggests that a laser cleaning technique may provide a solution to prevent or minimize wheel loading and maintain a sharp wheel surface. Research on utilizing lasers to dress grinding wheel surfaces has been reported [2,4,10]. The results demonstrated that the use of a laser can be an option used in grinding wheel dressing. However, laser dressing did not show a great advantage over conventional dressing. The possible reason is that a high powered laser not only removes wheel bonding materials but also damages the abrasive grains which then causes higher grinding forces to occur and results in higher wheel wear. A large degree of wheel wear is not acceptable for super-abrasive grinding wheels. It should be noted that the research has also pointed out that clogged chips can be removed through evaporation caused by laser radiation [2,4]. This suggests that a laser cleaning technique may be used to prevent, or minimize, wheel loading and maintain the sharp wheel surface created by techniques such as touch dressing. By continually irradiating the loaded wheel surface with a particular degree of laser energy, it is possible to remove clogged chips without deteriorating the wheel surface. 2. Experimental 2.1. Materials Chromium-doped alumina was selected as the abrasive grain material as it is commonly used for grinding tough engineering materials such as microalloyed steels, and is used in a wide variety of industrial applications such as roll grinding, camshaft grinding, crankshaft grinding, and other automotive component grinding operations [3]. The abrasive grains were bonded together in the form of a cylindrical grinding wheel by mixing a bonding system with the abrasive grains. The grains were coated in dextrin and mixed with water to form a sticky, wet mass. The mass of bonding ingredients was pressed to shape at a pre-determined density such that a porosity level of about 40% volume was achieved. After manufacturing the wheels by firing the bond constituents in a high temperature furnace, the wheels were graded as having a K-grade (medium hardness), six-structure (relatively open wheel), and possessing a single grain size of approximately 220 m diameter (60 mesh grain size). The relative proportions of grain, bonding system, and porosity, by volume were 50, 9.5, and 40.5, respectively. The workpiece materials to be ground during the grinding experiments to test the effectiveness of laser dressing includes a number of steel materials such as: high speed steels M1, M2, M15, M42, T1, T4, and T15; die steels D2 and D3; oil-hardening steel O1; and a powder metallurgical material known as ASP23.
19
Fig. 1. Arrangement of laser cleaning procedure.
2.2. Laser cleaning A carbon dioxide laser was used to explore the effects of laser irradiation on the surface of the grinding wheel and on clogged metal chips. The laser output from the machine can be continuous wave, or pulsed. For the investigation of the effects of laser energy transfer the pulsed output mode was used. As shown in Fig. 1, a laser beam passes through a convex lens and irradiates a moving grinding wheel surface. The wheel travel speed is 25.4 mm/s. A focused laser with high power density will cut through the wheel removing both cutting grains and clogged chips. Such a case can be considered as a laser dressing procedure. For laser cleaning, however, the aim is for the laser beam to remove clogged metal chips without damaging the wheel surface. This may be achieved by controlling laser irradiation on the wheel surface. Two important parameters noted in this study are the laser beam energy flux, and the duration of laser irradiation. By adjusting the focal offset, l, between the lens focus point and the wheel surface, the laser irradiation energy flux can be controlled. A large focal offset will provide a large laser spot on the surface of the wheel. Hence, the laser power flux will be lower. Control of laser pulse duration and pulse frequency can also adjust the level of laser irradiation power. Longer laser pulse durations and higher pulse frequencies will impart more laser energy onto the wheel surface. Grinding wheel surfaces were examined using a microscope.
2.3. Laser dressing The edges of the grinding wheels were dressed with a laser using various power intensities. A 2.5 kW Hobart continuous wave Nd:YAG laser equipped with a fiber-optic beam delivery system was used for dressing the Al2 O3 grinding wheel. The laser beam was focused at a height of 0.5 mm above the surface of the wheel. The lenses within the output-coupling module of the fiber optic delivery system were configured to provide 3.5 mm × 600 m line in spatial distribution onto the sample surface. Such configuration provides rapid processing speed and limits the overlap between laser strokes to 20%. Laser power intensities of 500, 750 and 1000 W were employed. Dressing of the entire surface was done by scanning the laser beam in parallel tracks at a speed of 50 cm/min. In order to analyze the dressed samples in detail using microstructural and other techniques, small pieces were cut out of the dressed grinding wheel. A slow speed diamond cutter was used for this purpose. Microstructural analysis of the samples was performed using scanning electron microscopy (SEM). To minimize charging during SEM analysis, a low vacuum (∼30 Pa air) was maintained in the SEM chamber.
2.4. Grinding experiments Grinding experiments were designed so as to replicate the conditions under which bond fracture is the dominant mechanism of wear [11]. The method used to measure wear of laser dressed grinding wheels was the ‘razor blade’ method. The method involves grinding a sample workpiece that is less narrow than the workpiece material. A groove is worn into the profile of the grinding wheel that was measured with reference to the non-grinding portion of the wheel using the razor blade. The blade was lowered into position with the grinding wheel just touching the blade. The table was then traversed until the wear profile was replicated onto the razor blade.
20
M.J. Jackson et al. / Journal of Materials Processing Technology 185 (2007) 17–23
The wear parameter used for quantifying the effectiveness of the grinding process is referred to as the grinding ratio (G-ratio) G=
Vw Vs
(3)
Eq. (3) is simply the ratio of the volume of the workpiece removed, Vw , and the volume of the grinding wheel removed, Vs . To replicate the conditions of bond fracture during grinding [11], the depth of cut was set at 10 m, and the machine table and grinding wheel speeds set at 0.2 and 30 m/s, respectively. The grinding system used for the experimental work was based on the Abwood 5025 surface-grinding machine retrofitted to measure grinding temperature and forces using fast thermocouples and piezoelectric sensors. The maximum spindle speed was rated at 3400 rpm with a maximum spindle power of 10 kW. Laser dressed samples were compared to conventionally dressed samples that used a single point diamond dresser. The details of the conventional dressing and grinding procedures are described by Jackson [12].
3. Results and discussion 3.1. Laser cleaning
Fig. 3. Grinding wheel surface after laser irradiation at the specified operating conditions.
A grinding trial was carried out on a cutting tool grinding machine to obtain a loaded wheel surface for further laser cleaning experiments. A vitrified grinding wheel was used to grind samples of inconel. Fig. 2 shows inconel material (bright shaded areas) attached to the surface of the grinding wheel after a sample had been ground. Grains of grinding wheel appear to be transparent using reflected light. A selection of control parameters is used for the laser cleaning experiments. Fig. 3 shows the effect of laser cleaning on the loaded wheel surface, where laser pulse duration is 20 × 10−5 s and laser pulse frequency is 19.9 Hz. When the laser is irradiated on the surface of the wheel, the metal chips absorbed the laser energy and melted. The chips also possess a dark color due to oxidation. For the focal offset length of l = 10 mm, most metal chips were melted by the laser beam and the wheel material was not damaged. However, a large chip could still be seen clogged in the wheel surface. By shortening the focal offset distance to 5 mm, the higher density laser power flux removed the chips but also damaged the wheel surface, as shown in Fig. 4. Results presented in Figs. 3 and 4 indicate that effective laser cleaning can only be achieved by applying a suitable laser power flux density.
The laser pulse frequency can also control the level of laser irradiance. Fig. 5 shows that effective laser cleaning was achieved when the laser pulse frequency was set to 98 Hz. The molten chips became dark spheres, scattering under the surface, and would tend to be expelled from the wheel when it was rotated at high speed. This demonstrated that wheel cleaning could be achieved through chip fusion and evaporation. However, when the laser pulse frequency was increased to 192 Hz, damage to the wheel surface material could clearly be seen, as shown in Fig. 6. This means that too much energy was irradiated on to the wheel surface. Some interesting observations were discovered during the experiments. Clogged chips remained in the wheel if a longer laser pulse length was used but with a low power flux. This was due to the fact that the heating process is gradual. The long pulse duration could however, ultimately lead to a high overall temperature on the wheel surface, which may harm the structure of
Fig. 2. Enlarged view of the surface of a loaded grinding wheel.
Fig. 4. Damaged wheel surface after laser cleaning at the specified conditions.
M.J. Jackson et al. / Journal of Materials Processing Technology 185 (2007) 17–23
Fig. 5. Effective laser cleaning using an adequate laser power flux.
the wheel. Some metal chips clogged in surface of the wheel are difficult to clean with low laser power flux, even using longer pulse durations. Longer laser irradiation may cause wheel damage rather than contribute to chip removal. When a laser beam irradiates the wheel surface, it may be regarded that energy flows in one direction in a semi-infinite body. If there is no convection or heat generation the basic equation shown by Eq. (1) is applicable. If it is assumed that the laser power flux input into the wheel is q, with no radiant heat loss, or melting, then a solution is shown using Eq. (2). Eq. (2) indicates that both q and t contribute to the elevation of the surface temperature. The depth to which laser energy penetrated into the wheel surface is constrained by the duration of the laser irradiation. Increasing irradiation time will allow laser energy to penetrate deeper into the wheel in order to raise the wheel’s temperature. For the purpose of laser cleaning, higher surface temperature rise is desirable for the chip removal. However, elevation of temperature under the wheel surface may damage the wheel structure that should be prevented. There-
Fig. 6. Higher irradiation frequency provides more energy on the surface of the grinding wheel.
21
fore, high power flux and short irradiation laser pulses are likely to be required for laser cleaning purposes. Chip fusion could play an important role in wheel cleaning. The melting points of materials may provide an indication to achieve effective cleaning. A nickel alloy’s melting point is less than 1500 ◦ C while that of Al2 O3 is higher than 2000 ◦ C. This indicates that there is a temperature range that allows chips to be removed without damaging the wheel material. In addition to the melting point, other differential factors relating to the interaction of the laser beam with the constituent parts of the loaded wheel should also be considered. The wheel grain, bonding material and the embedded metal, in addition to their melting points, will have other properties that vary between them. These properties include optical reflectivity, thermal and optical conductivity, and specific heats capacities. Understanding in detail how the laser reacts to these variations can be exploited to optimize the cleaning process. Therefore, selection of correct parameters for laser irradiation is very important for an effective and reliable cleaning process. 3.2. Laser dressing The dressed wheel surface has relatively large particles (100–150 m) with considerable porosity in between. The particles are irregular in shape with a few bonding bridges between them. Within the range of laser power employed, the sample surface underwent a transformation ranging from liquid–solid to solid–solid. Depending upon the laser processing parameters used, the high thermal energy produced during laser processing caused melting and/or vaporization of the grinding wheel material on surface. Thus both melting (followed by re-solidification) and/or vaporization resulted in modification of surface topography. The newly formed re-solidified layer had grains with a modified geometry, and these grains are characterized by multifaceted surfaces. Also, it was observed that laser dressing reduced the porosity on the region near to the surface of the grinding wheel. Complete densification of the surface was not observed, which is a desired feature of the surface of grinding wheels. This is because the grinding grains at the surface of wheel should break away once they become blunt, and expose the sharp particles that are present immediately below the surface, thus maintaining a high grinding efficiency at all the times (high G-ratio). A wide distribution of the size of the grinding grains was seen, but the shape of the grains appeared to be regular or equi-axed with well-defined vertices and edges on each grinding grain. The vertices and edges can provide excellent cutting edges for improved grinding processes. Fig. 7 shows one such multifaceted grain on the surface of the sample. A change in the direction of dendritic growth can be seen across the cutting edges and at the vertex. Owing to the fact that the laser is a concentrated high-energy source, and that the dressing process is fairly rapid, the solidification of the molten material on the surface is extremely fast. This results from near instantaneous solidification of molten material that produces a structure that is distinctly observed on the surface of the sample.
22
M.J. Jackson et al. / Journal of Materials Processing Technology 185 (2007) 17–23
Fig. 7. Morphological features on the surface of a laser dressed grinding wheel showing vertices and cutting edges.
Dendritic growth can be observed on the faces of the grinding grains. The refinement in microstructure due to laser dressing is not a characteristic feature of the surface alone. In fact, the refinement is on a much smaller scale in the layers below the surface. Owing to the fact that laser induced dressing is a rapid solidification process; thermal stresses are generated in the sample during dressing. These stresses are concentrated in the near-ridge region, and the results of which show themselves as cracks in the re-solidified layer. Different shapes of dendrites are present in this near ridge region. Detailed discussion of the interaction of the laser beam with the ceramic grinding wheel has been reported in Refs. [13,14]. 3.3. Grinding experiments for laser dressed wheels Laser dressed grinding wheel samples were compared to conventionally dressed grinding wheel structures. The workpiece materials selected for use in comparative experiments were chosen on the basis of their ability to grind under the stated experimental conditions, and to exploit the nature of the carbides contained within their structure. M1, M42, T1, and T4 were used to study the effects of cobalt on grinding ratio. M1, M2, and M15 were used to study the effects of vanadium on grinding ratio, whilst O1 and D3 are rich in chromium. The effect of carbon content is not compared since it is purposely controlled to allow the correct amount of complex carbides to precipitate during heat treatment. The presence of complex carbides had a significant effect on grinding ratio. The quality and hardness of the carbide particles can be classed as one of the most important factors in determining the grindability of tool steels. The combined effects of complex carbides can be expressed linearly to give an abrasive number, which is a weighted average of the Vickers’ hardness numbers of the complex carbides contained in the workpiece. Hence, An = Vmatrix Hmatrix + (Vcarbide Hcarbide ) (4) Where Vmatrix is the volume of the matrix structure, Hmatrix the hardness of the matrix structure, Vcarbide the volume of the car-
Fig. 8. Grinding ratio as a function of abrasive number for tool steel workpiece materials.
bides in the matrix, and Hcarbide is the hardness of the carbides in the matrix. The characteristics of the complex carbides in the workpiece materials are contained in Ref. [12]. Fig. 8 shows the comparative relationship between grinding ratio and abrasive number for diamond dressed and laser dressed grinding wheels. It can be seen that there is little difference in the way both wheels perform after dressing. From the results, it is clear that the effect of carbide content on grinding ratio has a more pronounced effect, although the mechanisms of grinding wheel wear are rather different when using optical energy to re-structure the grinding wheel rather than using diamond. It was observed that vitrified bonding bridge appearance and failure was markedly different in laser dressed grinding wheel samples than in conventionally dressed samples. Traditional vitrified bonding bridges are glassy structures that separate abrasive grains with a pronounced meniscus. Laser processed bonding bridges are replaced by blocky-shaped bonding bridges that are crystalline in nature but appear to possess such strength as to release abrasive grains at the same rate as conventionally dressed grinding wheels. Observations of the type of grinding wheel wear at high and low laser intensities appear to produce a wide variety of possibilities. In terms of bond bridge failure, low intensities tend to produce traditional fracture patterns in bond bridges (i.e. mirror, mist, and hackle patterns), whereas at high intensities the crystalline and blocky nature of the bonding between abrasive grains allows the grains to be held in place until the grain is dislodged due to crystal plane fracture in the blocky bonding bridge. The mechanism of grain fracture is a little more complicated to explain as it is not clear what is causing grain fracture to occur in some grinding wheel samples and not in others, at both high and low intensity levels. The lack of clarity is interesting as it appears to reveal itself as subtle changes in the grinding ratio, as shown in Fig. 8. The lower grinding ratio values are
M.J. Jackson et al. / Journal of Materials Processing Technology 185 (2007) 17–23
dominated by bond bridge failure, whilst higher grinding ratios appear to be governed by abrasive grain fracture, which is the preferred mode of wear of grinding wheels from an economic viewpoint. 4. Conclusion A feasibility study has been carried out to investigate the application of laser cleaning technology to the grinding process. By irradiating a laser beam on to the surface of a loaded grinding wheel using carefully selected control parameters, it is possible to remove clogged metal chips without deteriorating the wheel’s cutting surface. Both fusion and evaporation of chips are important for the laser cleaning process. Suitable laser parameters are identified and it was suggested that high power flux and short irradiation laser pulses would be ideal for laser cleaning. The application of high power lasers can be developed into an important dressing tool, if the processing parameters are precisely controlled. The study so far gives very strong indication towards this, as a refinement in the microstructure is seen immediately after laser dressing. Cutting efficiency is increased as a result of the formation of sharp points on the surface of grinding grains, possessing a multifaceted structure with individual cutting edges on them. The nature of bonding between grinding grains after laser dressing allows the grains to adhere to the surface just long enough so that, after loosing their cutting efficiency, they appear to become detached from the surface of the grinding wheel and expose new sharp particles below them. X-ray analysis indicates that strong texture formation on the surface of the grinding wheel occurs after laser dressing. In some cases even dual texturing effects were observed. Laser processing has a very prominent effect on the morphological structure of the surface of alumina grinding wheels.
23
References [1] C. Zhang, Y.C. Shin, A novel laser-assisted truing and dressing technique for vitrified cBN wheels, Int. J. Mach. Tools Manu. 42 (2002) 825–835. [2] N. Ramesh Babu, V. Radhakrishnan, Y.V.G.S. Murti, Investigation on laser dressing of grinding wheels—Part I: preliminary study, J. Eng. Ind. 111 (1989) 244–252. [3] M.J. Jackson, B. Mills, Materials selection applied to vitrified alumina and CBN grinding wheels, J. Mater. Process. Technol. 108 (2000) 114–124. [4] N. Ramesh Babu, V. Radhakrishnan, Investigation on laser dressing of grinding wheels—Part II: grinding performance of a laser dressed aluminum oxide wheel, J. Eng. Ind. 111 (1989) 253–261. [5] X. Chen, W.B. Rowe, B. Mills, D.R. Allanson, Analysis and simulation of the grinding process. Part III: comparison with experiment, Int. J. Mach. Tools Manu. 36 (8) (1996) 897–906. [6] T.C. Buttery, A. Statham, J.B. Perical, Some effects of dressing on grinding performance, Wear 55 (1979) 195–219. [7] N. Ramesh Babu, V. Radhakrishnan, Influence of dressing feed on the performance of laser dressed Al2 O3 wheel in wet grinding, Int. J. Mach. Tools Manu. 35 (1995) 661–671. [8] X. Chen, W.B. Rowe, Analysis and simulation of the grinding process. Part II: mechanics of grinding, Int. J. Mach. Tools Manu. 36 (8) (1996) 883–896. [9] N. Morgan, Excimer lasers restore 14th century icons, Optoelectron. Laser Europe 7 (1993) 36–37. [10] E. Westk¨amper, Grinding assisted by Nd:YAG lasers, Ann. CIRP 44 (1) (1995) 317–320. [11] S.K. Bhattacharyya, V.L. Moffatt, Characteristics of micro wheel wear in grinding, Int. J. Mach. Tools Manu. 16 (1976) 325–334. [12] M.J. Jackson, A Study of Vitreous-Bonded Abrasive Materials. PhD Thesis, Liverpool John Moores University, UK, 1995 (Chapter 7—Effects of Workpiece and Bond Composition on Grinding Wheel Performance). [13] M.J. Jackson, G.M. Robinson, N.B. Dahotre, A. Khangar, R. Moss, Laser dressing of vitrified aluminum oxide grinding wheels, Brit. Ceram. Trans. 102 (6) (2003) 237–245. [14] A. Khangar, N.B. Dahotre, M.J. Jackson, G. Robinson, Laser dressing of alumina grinding wheels, in: N.B. Dahotre, R.J. Gaster, R.A. Hill, O.O. Popoola (Eds.), Heat Treating and Surface Engineering Congress and Exposition, ASM International, Materials Park, OH, 2003, pp. 423–426.
Laser cleaning and dressing of vitrified grinding wheels M.J. Jackson a,∗ , A. Khangar b , X. Chen c , G.M. Robinson a , V.C. Venkatesh d , N.B. Dahotre b a
Center for Advanced Manufacturing, Purdue University, College of Technology, 401 North Grant Sreet, West Lafayette, IN 47907-2021, USA Department of Materials Science and Engineering, University of Tennessee-Knoxville, 326 Dougherty Hall, 1512 Middle Drive, Knoxville, TN 37996, USA c School of Mechanical, Materials, and Manufacturing Engineering and Management, University of Nottingham, University Park, Nottingham NG7 2RD, UK d Department of Mechanical Engineering, University of Technology Malaysia, 81310 UTM Skudai, Johor Durul Ta’zim, Malaysia b
Abstract Grinding is an industrial process that produces engineering components with a desired surface finish. Prior to the development of continuous dressing operations, the grinding efficiency of vitrified grinding wheels deteriorates as the sharp cutting edges become blunt due to the formation of wear flats. Dressing is essentially a sharpening operation designed to generate a specific topography on the working surface of the grinding wheel. The use of high power lasers is being explored as a non-contact cleaning and dressing technique. In the present study, a high power laser was used to clean metal chips from the surface of the grinding wheel and to dress the wheel by causing phase transformations to occur on the surface of vitrified grinding wheel. Experimental results indicated that laser modified grinding wheels are comparable in performance to conventionally cleaned and dressed grinding wheels. © 2006 Elsevier B.V. All rights reserved. Keywords: Vitrified grinding wheels; Laser materials processing; Grinding; Tool steels; Aerospace materials
1. Introduction 1.1. Conventional dressing of grinding wheels Grinding wheels are used for grinding materials with poor machinability because of their longer life, high grinding efficiency, and dimensional stability [1]. The grinding wheel also gets loaded with metal chips during machining [2]. The existent grinding tools wear at a very high rate and also generate high frictional heat at the abrasive grain-metal chip contact region, which leads to surface and sub-surface damage in the ground components. Also, there is a loss of form tolerance and dimensional stability as grinding progresses over a large number of grinding passes [3]. Ideally, the abrasive particles on the surface of the grinding wheel should get automatically sharpened when worn out, by either entirely detaching from the wheel face or by fracture, thus exposing new particles with sharper cutting edges. However, in practice both geometric and functional characteristics of the grinding wheel have to be restored periodically by dressing [2]. Dressing is a sharpening operation designed to generate a particular surface topography on the cutting face of
∗
Corresponding author. Tel.: +1 765 494 0365; fax: +1 765 494 6219. E-mail address: [email protected] (M.J. Jackson).
0924-0136/$ – see front matter © 2006 Elsevier B.V. All rights reserved. doi:10.1016/j.jmatprotec.2006.03.109
the wheel. Dressing of the worn out grinding wheel surface is done by various means to modify its surface topography, which in turn improves grinding efficiency [4]. Dressing significantly affects the quality of the ground product, as characterized by its size and shape, surface roughness and integrity [5]. The conventional contact-type methods, like mechanical dressing using a diamond dresser, results in excessive grinding wheel material loss. The bond fracture and abrasive grain break-off, as a result of crushing of the dresser are the material removal mechanisms for contact type processes. In fact, only 10% of the wheel by volume is removed during actual grinding, while the rest is removed during dressing operations [6]. Mechanical dressing, though effective, also induces stresses and causes deep cracks and undercuts. These factors eventually cause loosening of the chunks of grains and reduce the number of effective cutting edges [2]. The conventional processes do not produce consistent grinding results due to dresser wear which affects the wheel surface topography and its performance in grinding. A worn-out dresser cannot produce sufficient protrusion of cutting grain edges [1]. To obtain consistent grinding outputs, either the dresser geometry has to be maintained or suitable dressing conditions corresponding to modified dresser geometry have to be selected, which are impossible in contact type processes and also complicated because of dresser wear [4]. Yet in order to obtain
18
M.J. Jackson et al. / Journal of Materials Processing Technology 185 (2007) 17–23
consistent grinding results, a dressing procedure, which is reproducible in nature, is essential. To achieve these requirements, methods have been developed to maintain an optimized grinding wheel. These methods include: slurry and steel roll application, pressurized jet dressing using abrasive slurries, and air jet abrasive dressing. In air jet dressing, the dressing rate is controlled in three ways: (a) by controlling the pressure of the mixed air, (b) by controlling the ejecting nozzle traverse speed, and (c) by controlling the number of traverse passes across the grinding wheel. Therefore, the dressing rate will control the optimum grain protrusion height. 1.2. Laser cleaning and dressing of grinding wheels High power lasers that are currently used as a non-contact type machining tool for various manufacturing applications such as welding, drilling, cutting, etc., can also be used as a non-contact type dressing tool. The salient features of a laser include high intensity fluence, directionality, and spatial coherence, which can be used to process hard and brittle materials efficiently. Laser induced thermal processing leads to effects such as melting, vaporization, and plasma formation on the material of the grinding wheel, which can be exploited during the dressing procedure. When a laser beam irradiates the surface of a grinding wheel, it may be considered that the energy flows in one direction in a semi-infinite body. If there is no convection, or heat generation, the basic equation governing the flow of heat is, ∂2 T 1 ∂T = 2 ∂z α ∂t
(1)
Where T is the temperature in the grinding wheel, z the depth from the wheel surface, α the thermal diffusivity of the grinding wheel, and t is the time after laser irradiation commences. If it is assumed that the laser power flux input into the wheel is q, with no radiant heat loss or melting, then the solution to Eq. (2) is, √ 2q z √ Tz,t = (2) α t ierfc K 2 αt Where K is the thermal conductivity of the grinding wheel material, and ierfc is the integral of the complementary error function. Eq. (2) indicates that both q and t contribute to elevating the temperature of the surface of the grinding wheel. The depth of laser energy penetrated into the wheel surface is constrained by the duration of the laser pulse. Increasing irradiation time will allow the laser energy to penetrate deeper into the grinding wheel. For the purpose of laser dressing, a higher temperature is needed in order to remove metal chips and re-shape individual grinding grains. During laser dressing, the wheel surface topography of the grinding wheel is modified by melting of the material and subsequent re-solidification of a portion of the molten layer. During the process, rapid heating and cooling induces cracks in the re-solidified layer. The microcracks help remove the re-melted layer during grinding after a few initial grinding strokes, which then exposes new cutting edges. In laser dressing, the grinding wheel is subjected to high power laser intensity, which produces
craters on the surface and also induces microcracks in the re-cast molten layer. By using different laser powers during dressing, the extent of the re-cast molten layer can be controlled and controlling the dressing feed can vary the surface topography. Laser dressing removes the wheel material by ablation of the bonding material to expose sharp grains. When the heating time and energy density is selected correctly, the vitrified bond (glass phase) of the wheel is softened and even melted thus facilitating the removal of the bonding material [1]. Dressing of the grinding wheel by laser generates welldefined grooves and tracks on the surface of the wheel. Essentially, laser dressing produces microcutting edges due to the formation of microcracks on the worn-out grains. When these craters are formed on the bond, the grits are loosened and subsequently removed due to insufficient volume of load surrounding the grain [4]. Laser dressing thus has the advantage of being a non-contact type process in which selective removal of clogged material alone is possible by appropriate focusing of the laser beam on the selected portions of the wheel surface. Also the material wastage in terms of the debris produced during dressing operation, can be substantially reduced by the use of a laser as a dressing tool. Thus environmentally benign grinding operations can be made possible, which will help reduce the problem of disposing contaminated grinding debris. There are several inherent advantages associated with the use of laser for dressing applications. Laser dressing is a very fast process and it can be easily automated. Also, selective removal of the clogged material alone is possible and desired surface structure (roughness, grain morphology and porosity) can be generated. Furthermore, consistent dressing conditions can be produced by the use of laser and this can help achieve grinding reproducibility. As the laser beam can be delivered using a fiber optic cable, remote dressing operation without discontinuation of the grinding process during laser dressing is possible. Thus the downtime in the grinding operation associated with conventional methods, can either be eliminated or substantially reduced in laser dressing. Earlier studies [2,4,7] used pulsed laser to compare the laser dressing process with conventional mechanical dressing methods. The pulsed laser powers used were of the order of 1–5 × 1010 W/m2 and most of the work concentrated on comparison of the grinding performance of laser dressed wheels with that of diamond dressed ones. Laser-assisted simultaneous truing and dressing has also been attempted, to overcome the problems associated with mechanical dressing [8]. Though useful, the studies did not deal with the nature of physical changes taking place in the grinding wheel during interaction with laser energy, which happens to be a fundamental aspect to contribute towards its dressing performance. Prior to any dressing operation, the grinding wheel tends to load with metal chips that need to be removed. Wheel loading is one of the most common problems in grinding operations, particularly for grinding aerospace materials. As grinding continues, removed chips may adhere in the spaces between abrasive grains and deteriorate the cutting ability of the grinding wheel. A common method used to prevent the wheel from loading is achieved by delivering a large amount of coolant to the grinding
M.J. Jackson et al. / Journal of Materials Processing Technology 185 (2007) 17–23
zone. However, this consumes huge amounts of energy in coolant delivery, especially for high speed grinding processes. Maintaining and disposing of coolant is also an environmental issue and the costs involved are substantial. Another method, which is often used, is to remove loaded chip materials by dressing the wheel periodically, to restore a sharp wheel surface. However, dressing of grinding wheels with diamond not only causes excessive wheel loss but also interrupts grinding during dressing. In addition, the dresser wears away with time due to its direct contact with the wheel surface. Frequent use of dressing wheels is also not acceptable for super-abrasive wheels whose cost is considerably more than conventional grinding wheels. Lasers have been successfully applied to material removal processes such as laser cutting, and drilling. Laser cleaning techniques have been used in removing pollution layers from valuable artifacts and without damaging the delicate patina of the substrate [9]. This suggests that a laser cleaning technique may provide a solution to prevent or minimize wheel loading and maintain a sharp wheel surface. Research on utilizing lasers to dress grinding wheel surfaces has been reported [2,4,10]. The results demonstrated that the use of a laser can be an option used in grinding wheel dressing. However, laser dressing did not show a great advantage over conventional dressing. The possible reason is that a high powered laser not only removes wheel bonding materials but also damages the abrasive grains which then causes higher grinding forces to occur and results in higher wheel wear. A large degree of wheel wear is not acceptable for super-abrasive grinding wheels. It should be noted that the research has also pointed out that clogged chips can be removed through evaporation caused by laser radiation [2,4]. This suggests that a laser cleaning technique may be used to prevent, or minimize, wheel loading and maintain the sharp wheel surface created by techniques such as touch dressing. By continually irradiating the loaded wheel surface with a particular degree of laser energy, it is possible to remove clogged chips without deteriorating the wheel surface. 2. Experimental 2.1. Materials Chromium-doped alumina was selected as the abrasive grain material as it is commonly used for grinding tough engineering materials such as microalloyed steels, and is used in a wide variety of industrial applications such as roll grinding, camshaft grinding, crankshaft grinding, and other automotive component grinding operations [3]. The abrasive grains were bonded together in the form of a cylindrical grinding wheel by mixing a bonding system with the abrasive grains. The grains were coated in dextrin and mixed with water to form a sticky, wet mass. The mass of bonding ingredients was pressed to shape at a pre-determined density such that a porosity level of about 40% volume was achieved. After manufacturing the wheels by firing the bond constituents in a high temperature furnace, the wheels were graded as having a K-grade (medium hardness), six-structure (relatively open wheel), and possessing a single grain size of approximately 220 m diameter (60 mesh grain size). The relative proportions of grain, bonding system, and porosity, by volume were 50, 9.5, and 40.5, respectively. The workpiece materials to be ground during the grinding experiments to test the effectiveness of laser dressing includes a number of steel materials such as: high speed steels M1, M2, M15, M42, T1, T4, and T15; die steels D2 and D3; oil-hardening steel O1; and a powder metallurgical material known as ASP23.
19
Fig. 1. Arrangement of laser cleaning procedure.
2.2. Laser cleaning A carbon dioxide laser was used to explore the effects of laser irradiation on the surface of the grinding wheel and on clogged metal chips. The laser output from the machine can be continuous wave, or pulsed. For the investigation of the effects of laser energy transfer the pulsed output mode was used. As shown in Fig. 1, a laser beam passes through a convex lens and irradiates a moving grinding wheel surface. The wheel travel speed is 25.4 mm/s. A focused laser with high power density will cut through the wheel removing both cutting grains and clogged chips. Such a case can be considered as a laser dressing procedure. For laser cleaning, however, the aim is for the laser beam to remove clogged metal chips without damaging the wheel surface. This may be achieved by controlling laser irradiation on the wheel surface. Two important parameters noted in this study are the laser beam energy flux, and the duration of laser irradiation. By adjusting the focal offset, l, between the lens focus point and the wheel surface, the laser irradiation energy flux can be controlled. A large focal offset will provide a large laser spot on the surface of the wheel. Hence, the laser power flux will be lower. Control of laser pulse duration and pulse frequency can also adjust the level of laser irradiation power. Longer laser pulse durations and higher pulse frequencies will impart more laser energy onto the wheel surface. Grinding wheel surfaces were examined using a microscope.
2.3. Laser dressing The edges of the grinding wheels were dressed with a laser using various power intensities. A 2.5 kW Hobart continuous wave Nd:YAG laser equipped with a fiber-optic beam delivery system was used for dressing the Al2 O3 grinding wheel. The laser beam was focused at a height of 0.5 mm above the surface of the wheel. The lenses within the output-coupling module of the fiber optic delivery system were configured to provide 3.5 mm × 600 m line in spatial distribution onto the sample surface. Such configuration provides rapid processing speed and limits the overlap between laser strokes to 20%. Laser power intensities of 500, 750 and 1000 W were employed. Dressing of the entire surface was done by scanning the laser beam in parallel tracks at a speed of 50 cm/min. In order to analyze the dressed samples in detail using microstructural and other techniques, small pieces were cut out of the dressed grinding wheel. A slow speed diamond cutter was used for this purpose. Microstructural analysis of the samples was performed using scanning electron microscopy (SEM). To minimize charging during SEM analysis, a low vacuum (∼30 Pa air) was maintained in the SEM chamber.
2.4. Grinding experiments Grinding experiments were designed so as to replicate the conditions under which bond fracture is the dominant mechanism of wear [11]. The method used to measure wear of laser dressed grinding wheels was the ‘razor blade’ method. The method involves grinding a sample workpiece that is less narrow than the workpiece material. A groove is worn into the profile of the grinding wheel that was measured with reference to the non-grinding portion of the wheel using the razor blade. The blade was lowered into position with the grinding wheel just touching the blade. The table was then traversed until the wear profile was replicated onto the razor blade.
20
M.J. Jackson et al. / Journal of Materials Processing Technology 185 (2007) 17–23
The wear parameter used for quantifying the effectiveness of the grinding process is referred to as the grinding ratio (G-ratio) G=
Vw Vs
(3)
Eq. (3) is simply the ratio of the volume of the workpiece removed, Vw , and the volume of the grinding wheel removed, Vs . To replicate the conditions of bond fracture during grinding [11], the depth of cut was set at 10 m, and the machine table and grinding wheel speeds set at 0.2 and 30 m/s, respectively. The grinding system used for the experimental work was based on the Abwood 5025 surface-grinding machine retrofitted to measure grinding temperature and forces using fast thermocouples and piezoelectric sensors. The maximum spindle speed was rated at 3400 rpm with a maximum spindle power of 10 kW. Laser dressed samples were compared to conventionally dressed samples that used a single point diamond dresser. The details of the conventional dressing and grinding procedures are described by Jackson [12].
3. Results and discussion 3.1. Laser cleaning
Fig. 3. Grinding wheel surface after laser irradiation at the specified operating conditions.
A grinding trial was carried out on a cutting tool grinding machine to obtain a loaded wheel surface for further laser cleaning experiments. A vitrified grinding wheel was used to grind samples of inconel. Fig. 2 shows inconel material (bright shaded areas) attached to the surface of the grinding wheel after a sample had been ground. Grains of grinding wheel appear to be transparent using reflected light. A selection of control parameters is used for the laser cleaning experiments. Fig. 3 shows the effect of laser cleaning on the loaded wheel surface, where laser pulse duration is 20 × 10−5 s and laser pulse frequency is 19.9 Hz. When the laser is irradiated on the surface of the wheel, the metal chips absorbed the laser energy and melted. The chips also possess a dark color due to oxidation. For the focal offset length of l = 10 mm, most metal chips were melted by the laser beam and the wheel material was not damaged. However, a large chip could still be seen clogged in the wheel surface. By shortening the focal offset distance to 5 mm, the higher density laser power flux removed the chips but also damaged the wheel surface, as shown in Fig. 4. Results presented in Figs. 3 and 4 indicate that effective laser cleaning can only be achieved by applying a suitable laser power flux density.
The laser pulse frequency can also control the level of laser irradiance. Fig. 5 shows that effective laser cleaning was achieved when the laser pulse frequency was set to 98 Hz. The molten chips became dark spheres, scattering under the surface, and would tend to be expelled from the wheel when it was rotated at high speed. This demonstrated that wheel cleaning could be achieved through chip fusion and evaporation. However, when the laser pulse frequency was increased to 192 Hz, damage to the wheel surface material could clearly be seen, as shown in Fig. 6. This means that too much energy was irradiated on to the wheel surface. Some interesting observations were discovered during the experiments. Clogged chips remained in the wheel if a longer laser pulse length was used but with a low power flux. This was due to the fact that the heating process is gradual. The long pulse duration could however, ultimately lead to a high overall temperature on the wheel surface, which may harm the structure of
Fig. 2. Enlarged view of the surface of a loaded grinding wheel.
Fig. 4. Damaged wheel surface after laser cleaning at the specified conditions.
M.J. Jackson et al. / Journal of Materials Processing Technology 185 (2007) 17–23
Fig. 5. Effective laser cleaning using an adequate laser power flux.
the wheel. Some metal chips clogged in surface of the wheel are difficult to clean with low laser power flux, even using longer pulse durations. Longer laser irradiation may cause wheel damage rather than contribute to chip removal. When a laser beam irradiates the wheel surface, it may be regarded that energy flows in one direction in a semi-infinite body. If there is no convection or heat generation the basic equation shown by Eq. (1) is applicable. If it is assumed that the laser power flux input into the wheel is q, with no radiant heat loss, or melting, then a solution is shown using Eq. (2). Eq. (2) indicates that both q and t contribute to the elevation of the surface temperature. The depth to which laser energy penetrated into the wheel surface is constrained by the duration of the laser irradiation. Increasing irradiation time will allow laser energy to penetrate deeper into the wheel in order to raise the wheel’s temperature. For the purpose of laser cleaning, higher surface temperature rise is desirable for the chip removal. However, elevation of temperature under the wheel surface may damage the wheel structure that should be prevented. There-
Fig. 6. Higher irradiation frequency provides more energy on the surface of the grinding wheel.
21
fore, high power flux and short irradiation laser pulses are likely to be required for laser cleaning purposes. Chip fusion could play an important role in wheel cleaning. The melting points of materials may provide an indication to achieve effective cleaning. A nickel alloy’s melting point is less than 1500 ◦ C while that of Al2 O3 is higher than 2000 ◦ C. This indicates that there is a temperature range that allows chips to be removed without damaging the wheel material. In addition to the melting point, other differential factors relating to the interaction of the laser beam with the constituent parts of the loaded wheel should also be considered. The wheel grain, bonding material and the embedded metal, in addition to their melting points, will have other properties that vary between them. These properties include optical reflectivity, thermal and optical conductivity, and specific heats capacities. Understanding in detail how the laser reacts to these variations can be exploited to optimize the cleaning process. Therefore, selection of correct parameters for laser irradiation is very important for an effective and reliable cleaning process. 3.2. Laser dressing The dressed wheel surface has relatively large particles (100–150 m) with considerable porosity in between. The particles are irregular in shape with a few bonding bridges between them. Within the range of laser power employed, the sample surface underwent a transformation ranging from liquid–solid to solid–solid. Depending upon the laser processing parameters used, the high thermal energy produced during laser processing caused melting and/or vaporization of the grinding wheel material on surface. Thus both melting (followed by re-solidification) and/or vaporization resulted in modification of surface topography. The newly formed re-solidified layer had grains with a modified geometry, and these grains are characterized by multifaceted surfaces. Also, it was observed that laser dressing reduced the porosity on the region near to the surface of the grinding wheel. Complete densification of the surface was not observed, which is a desired feature of the surface of grinding wheels. This is because the grinding grains at the surface of wheel should break away once they become blunt, and expose the sharp particles that are present immediately below the surface, thus maintaining a high grinding efficiency at all the times (high G-ratio). A wide distribution of the size of the grinding grains was seen, but the shape of the grains appeared to be regular or equi-axed with well-defined vertices and edges on each grinding grain. The vertices and edges can provide excellent cutting edges for improved grinding processes. Fig. 7 shows one such multifaceted grain on the surface of the sample. A change in the direction of dendritic growth can be seen across the cutting edges and at the vertex. Owing to the fact that the laser is a concentrated high-energy source, and that the dressing process is fairly rapid, the solidification of the molten material on the surface is extremely fast. This results from near instantaneous solidification of molten material that produces a structure that is distinctly observed on the surface of the sample.
22
M.J. Jackson et al. / Journal of Materials Processing Technology 185 (2007) 17–23
Fig. 7. Morphological features on the surface of a laser dressed grinding wheel showing vertices and cutting edges.
Dendritic growth can be observed on the faces of the grinding grains. The refinement in microstructure due to laser dressing is not a characteristic feature of the surface alone. In fact, the refinement is on a much smaller scale in the layers below the surface. Owing to the fact that laser induced dressing is a rapid solidification process; thermal stresses are generated in the sample during dressing. These stresses are concentrated in the near-ridge region, and the results of which show themselves as cracks in the re-solidified layer. Different shapes of dendrites are present in this near ridge region. Detailed discussion of the interaction of the laser beam with the ceramic grinding wheel has been reported in Refs. [13,14]. 3.3. Grinding experiments for laser dressed wheels Laser dressed grinding wheel samples were compared to conventionally dressed grinding wheel structures. The workpiece materials selected for use in comparative experiments were chosen on the basis of their ability to grind under the stated experimental conditions, and to exploit the nature of the carbides contained within their structure. M1, M42, T1, and T4 were used to study the effects of cobalt on grinding ratio. M1, M2, and M15 were used to study the effects of vanadium on grinding ratio, whilst O1 and D3 are rich in chromium. The effect of carbon content is not compared since it is purposely controlled to allow the correct amount of complex carbides to precipitate during heat treatment. The presence of complex carbides had a significant effect on grinding ratio. The quality and hardness of the carbide particles can be classed as one of the most important factors in determining the grindability of tool steels. The combined effects of complex carbides can be expressed linearly to give an abrasive number, which is a weighted average of the Vickers’ hardness numbers of the complex carbides contained in the workpiece. Hence, An = Vmatrix Hmatrix + (Vcarbide Hcarbide ) (4) Where Vmatrix is the volume of the matrix structure, Hmatrix the hardness of the matrix structure, Vcarbide the volume of the car-
Fig. 8. Grinding ratio as a function of abrasive number for tool steel workpiece materials.
bides in the matrix, and Hcarbide is the hardness of the carbides in the matrix. The characteristics of the complex carbides in the workpiece materials are contained in Ref. [12]. Fig. 8 shows the comparative relationship between grinding ratio and abrasive number for diamond dressed and laser dressed grinding wheels. It can be seen that there is little difference in the way both wheels perform after dressing. From the results, it is clear that the effect of carbide content on grinding ratio has a more pronounced effect, although the mechanisms of grinding wheel wear are rather different when using optical energy to re-structure the grinding wheel rather than using diamond. It was observed that vitrified bonding bridge appearance and failure was markedly different in laser dressed grinding wheel samples than in conventionally dressed samples. Traditional vitrified bonding bridges are glassy structures that separate abrasive grains with a pronounced meniscus. Laser processed bonding bridges are replaced by blocky-shaped bonding bridges that are crystalline in nature but appear to possess such strength as to release abrasive grains at the same rate as conventionally dressed grinding wheels. Observations of the type of grinding wheel wear at high and low laser intensities appear to produce a wide variety of possibilities. In terms of bond bridge failure, low intensities tend to produce traditional fracture patterns in bond bridges (i.e. mirror, mist, and hackle patterns), whereas at high intensities the crystalline and blocky nature of the bonding between abrasive grains allows the grains to be held in place until the grain is dislodged due to crystal plane fracture in the blocky bonding bridge. The mechanism of grain fracture is a little more complicated to explain as it is not clear what is causing grain fracture to occur in some grinding wheel samples and not in others, at both high and low intensity levels. The lack of clarity is interesting as it appears to reveal itself as subtle changes in the grinding ratio, as shown in Fig. 8. The lower grinding ratio values are
M.J. Jackson et al. / Journal of Materials Processing Technology 185 (2007) 17–23
dominated by bond bridge failure, whilst higher grinding ratios appear to be governed by abrasive grain fracture, which is the preferred mode of wear of grinding wheels from an economic viewpoint. 4. Conclusion A feasibility study has been carried out to investigate the application of laser cleaning technology to the grinding process. By irradiating a laser beam on to the surface of a loaded grinding wheel using carefully selected control parameters, it is possible to remove clogged metal chips without deteriorating the wheel’s cutting surface. Both fusion and evaporation of chips are important for the laser cleaning process. Suitable laser parameters are identified and it was suggested that high power flux and short irradiation laser pulses would be ideal for laser cleaning. The application of high power lasers can be developed into an important dressing tool, if the processing parameters are precisely controlled. The study so far gives very strong indication towards this, as a refinement in the microstructure is seen immediately after laser dressing. Cutting efficiency is increased as a result of the formation of sharp points on the surface of grinding grains, possessing a multifaceted structure with individual cutting edges on them. The nature of bonding between grinding grains after laser dressing allows the grains to adhere to the surface just long enough so that, after loosing their cutting efficiency, they appear to become detached from the surface of the grinding wheel and expose new sharp particles below them. X-ray analysis indicates that strong texture formation on the surface of the grinding wheel occurs after laser dressing. In some cases even dual texturing effects were observed. Laser processing has a very prominent effect on the morphological structure of the surface of alumina grinding wheels.
23
References [1] C. Zhang, Y.C. Shin, A novel laser-assisted truing and dressing technique for vitrified cBN wheels, Int. J. Mach. Tools Manu. 42 (2002) 825–835. [2] N. Ramesh Babu, V. Radhakrishnan, Y.V.G.S. Murti, Investigation on laser dressing of grinding wheels—Part I: preliminary study, J. Eng. Ind. 111 (1989) 244–252. [3] M.J. Jackson, B. Mills, Materials selection applied to vitrified alumina and CBN grinding wheels, J. Mater. Process. Technol. 108 (2000) 114–124. [4] N. Ramesh Babu, V. Radhakrishnan, Investigation on laser dressing of grinding wheels—Part II: grinding performance of a laser dressed aluminum oxide wheel, J. Eng. Ind. 111 (1989) 253–261. [5] X. Chen, W.B. Rowe, B. Mills, D.R. Allanson, Analysis and simulation of the grinding process. Part III: comparison with experiment, Int. J. Mach. Tools Manu. 36 (8) (1996) 897–906. [6] T.C. Buttery, A. Statham, J.B. Perical, Some effects of dressing on grinding performance, Wear 55 (1979) 195–219. [7] N. Ramesh Babu, V. Radhakrishnan, Influence of dressing feed on the performance of laser dressed Al2 O3 wheel in wet grinding, Int. J. Mach. Tools Manu. 35 (1995) 661–671. [8] X. Chen, W.B. Rowe, Analysis and simulation of the grinding process. Part II: mechanics of grinding, Int. J. Mach. Tools Manu. 36 (8) (1996) 883–896. [9] N. Morgan, Excimer lasers restore 14th century icons, Optoelectron. Laser Europe 7 (1993) 36–37. [10] E. Westk¨amper, Grinding assisted by Nd:YAG lasers, Ann. CIRP 44 (1) (1995) 317–320. [11] S.K. Bhattacharyya, V.L. Moffatt, Characteristics of micro wheel wear in grinding, Int. J. Mach. Tools Manu. 16 (1976) 325–334. [12] M.J. Jackson, A Study of Vitreous-Bonded Abrasive Materials. PhD Thesis, Liverpool John Moores University, UK, 1995 (Chapter 7—Effects of Workpiece and Bond Composition on Grinding Wheel Performance). [13] M.J. Jackson, G.M. Robinson, N.B. Dahotre, A. Khangar, R. Moss, Laser dressing of vitrified aluminum oxide grinding wheels, Brit. Ceram. Trans. 102 (6) (2003) 237–245. [14] A. Khangar, N.B. Dahotre, M.J. Jackson, G. Robinson, Laser dressing of alumina grinding wheels, in: N.B. Dahotre, R.J. Gaster, R.A. Hill, O.O. Popoola (Eds.), Heat Treating and Surface Engineering Congress and Exposition, ASM International, Materials Park, OH, 2003, pp. 423–426.