Investigation on minimum quantity lubricant-MQL grinding of 100Cr6 hardened steel using different abrasive and coolant–lubricant types

Investigation on minimum quantity lubricant-MQL grinding of 100Cr6 hardened steel using different abrasive and coolant–lubricant types

ARTICLE IN PRESS International Journal of Machine Tools & Manufacture 50 (2010) 698–708 Contents lists available at ScienceDirect International Jour...

6MB Sizes 2 Downloads 116 Views

ARTICLE IN PRESS International Journal of Machine Tools & Manufacture 50 (2010) 698–708

Contents lists available at ScienceDirect

International Journal of Machine Tools & Manufacture journal homepage: www.elsevier.com/locate/ijmactool

Investigation on minimum quantity lubricant-MQL grinding of 100Cr6 hardened steel using different abrasive and coolant–lubricant types T. Tawakoli a, M.J. Hadad a,b,n, M.H. Sadeghi b a b

Institute of Grinding and Precision Technology (KSF), Furtwangen University, 78054 VS-Schwenningen, Germany CAD/CAM Laboratory, Manufacturing Engineering Division, School of Engineering, Tarbiat Modares University, Tehran, Iran

a r t i c l e in f o

a b s t r a c t

Article history: Received 24 January 2010 Received in revised form 18 April 2010 Accepted 21 April 2010 Available online 27 April 2010

Large quantities of coolant–lubricants are still widely used in the metal working industry, generating high consumption and discard costs and impacting the environment. Alternatives to current practices are getting more serious consideration in response to environmental and operational cost pressures. In the grinding process, promising alternatives to conventional dry and fluid coolant applications are minimum quantity lubrication (MQL) or near dry grinding process. Despite several researches, there have been a few investigations about the influence of different types of coolant–lubricants and grinding wheels on the process results. The current study aims to show the effects of the above parameters on grinding performance such as grinding forces and surface quality. The tests have been performed in presence of fluid, air jet and eleven types of coolant–lubricants, as well as, in dry condition. The grinding wheels employed in this study were vitrified bond corundum, resin bond corundum and vitrified bond SG wheels. The results indicate that SG wheels and MQL oils have potential for the development of the MQL process in comparison to vitrified and resin bond corundums and water miscible oils. Also, the lowest thermal damages, material side flow on the ground surface and wheel loading were generated by using the SG grinding wheel in MQL grinding process. & 2010 Elsevier Ltd. All rights reserved.

Keywords: Minimum quantity lubrication (MQL) Grinding wheel Oil mist lubrication Coolant–lubricant Grinding forces Surface quality

1. Introduction Grinding is mostly a final process on the workpiece that the dimensional and form accuracy as well as surface quality is very important. Therefore the negative effect of high temperature on these parameters should be prevented. In spite of many advantages of the use of cutting fluids in the machining processes, they have serious disadvantages, such as ecological and economical problems, which have guided research works in the last decades to reduce or even eliminate the use of metal cutting fluids [1–7]. One attractive alternative for dry and fluid grinding processes is MQL grinding. This process uses a minimum quantity of lubrication and is referred to as near dry grinding. In this process aerosols are oil droplets dispersed in a jet of air, oil droplets carried by the air fly directly to the tool working zone, providing the needed cooling and lubricating actions [8,9]. Tawakoli et al. [10] investigated the effects of the workpiece material hardness and grinding parameters on the MQL grinding process. Based on the results of their investigations, n Corresponding author at: Institute of Grinding and Precision Technology (KSF), Furtwangen University, 78054 VS-Schwenningen, Germany. Tel.: + 49 7720 307 4294; fax: + 49 7720 307 4208. E-mail address: [email protected] (M.J. Hadad).

0890-6955/$ - see front matter & 2010 Elsevier Ltd. All rights reserved. doi:10.1016/j.ijmachtools.2010.04.009

significant improvement can be achieved by MQL grinding of hardened steel in comparison to dry grinding process. Brunner [11] showed that the MQL grinding of 16MnCr5 (SAE-5115) with 4 ml/min ester oil, as compared to 11 ml/min mineral oil, reduces process normal and tangential forces to one third, but increases surface roughness by up to 50%. Investigations by Brinksmeier et al. [12] confirmed these results and showed additionally that the type of coolant–lubricant can also considerably influence the MQL process. In order to achieve a finer surface quality and avoid thermal damage to the workpiece, the chip formation mechanism has to be improved so that both the friction forces and thermal partition to the workpiece reduces. Hence, the influences of the wheel and coolant–lubricant characteristics on the MQL grinding process have to be studied from the grinding forces, chip formation and temperature generation stand points. Literature review shows the lack of study on the effects of grinding wheel and MQL oil type in minimum quantity lubrication grinding process. In this paper experiments conducted under different Al2O3 grinding wheels and coolant– lubricant types, showed the performance of the MQL grinding operation based on an evaluation of grinding forces and surface quality.

ARTICLE IN PRESS T. Tawakoli et al. / International Journal of Machine Tools & Manufacture 50 (2010) 698–708

2. Experimental procedure The experimental setup is summarized in Table 1. In order to compare the effects of abrasive and coolant–lubricant types on MQL performances, three AL2O3 grinding wheels with the same grain size and porosity were used (Table 1). The grinding wheels employed in this study were vitrified bond corundum, resin bond corundum and vitrified bond SG wheels. The vitrified bond corundum and SG wheels had hardness (I), while the resin bond corundum had the hardness degree (E). To investigate the effects of MQL fluid type, eleven types of oil and water miscible grinding fluids with different viscosities (Table 2) as well as conventional dry, fluid and air jet grinding conditions were used. The grinding wheels were dressed before each experiment with the conditions shown in Table 1. Surface grinding tests were done through the 8 mm width for 100Cr6 hardened steel using ELB micro-cut AC8 CNC universal surface grinding machine. The total depth of material removal in each grinding test was 9.0 mm (300 grinding passes).The equipment utilized to control the minimum quantity of lubricant (MQL) was Accu-Lube system in which an oil supply pump system is used. In this system, the compressed air and lubricant flow can be adjusted separately and mixed in the special nozzle (with nominal diameter 3 mm) to make microdroplets of cutting oil flying to the cutting zone by compressed air. The workpiece roughness was measured by Hommel Tester T-1000 (mobile roughness measurement) with a cut-off length of 0.8 mm (according to DIN EN ISO 3274:1998). At the end of each test (after the 300th pass), Rz and Ra across the grinding direction were measured at five different points of ground surface. The grinding force components were recorded using a piezo-electric transducer based dynamometer (type Kistler 9255B) positioned under the workpiece clamping device. Chips, surface morphology and grinding wheel surfaces were observed using a digital microscope (Keyence: VHX) which possesses a maximum magnification of 1000 times.

699

region of strong adhesion. In other words, if the lubricant is unable to access to the whole of the interface between the grain and the workpiece surface, then at least it is necessary to limit the region of adhesion and so to reduce the friction forces exhibited at the grain-workpiece boundary. It is possible to consider that the oil droplets access the contact zone through lubricant sources, namely the fractured grooves on the grain wear flat area and wheel pores (spaces between grains on the wheel or porosity). The rate of transportation of the oil droplets within these lubricant sources can help the lubrication of the grinding zone. Also, the pores on the surface of the wheel entrain and pump the oil mist through the grinding contact to cool the cutting zone [1]. Grain fractured grooves and the wheel pores in the contact zone as lubricant sources are shown in Fig. 1. Because the grain fractures are developed by dressing process and soon after the beginning of the grinding process, the existence of the lubricant sources between the grain wear flat area and the workpiece surface could be always taken into account. Lubricating action of the MQL oil mist requires penetration of the oil droplets into the lubricant sources. Therefore, the penetration time of the oil droplets into the lubricant source must be less than the passing time of the lubricant source. The passing time of the lubricant source is the time that lubricant source passes from the leading edge in front of the nozzle to the contact zone. Therefore, the penetration time can be defined as tpen ¼ ls/Voilmist, and the lubricant source passing time can be calculated from tpass ¼ls/Vc. ls is the length of the lubricant source, Voilmist indicates the average velocity of the oil droplet penetration into the lubricant

Table 2 MQL coolant–lubricants used in this study.

3. Lubrication mechanism in MQL grinding process The physical and chemical properties of the coolant–lubricants determine their effectiveness in the grinding process. By reaction of the coolant–lubricant with the workpiece material in the contact zone, intermediate layers can emerge to separate the surfaces and reduce friction [13]. A successful coolant–lubricant might be thought to form a low shear strength layer between the grain wear flat face and the workpiece surface to eliminate the

MQL coolant–lubricant

viscosity (mm2/s) at 40 1C

MQL oil 1 (based on mineral oil) MQL oil 2 (based on fat alcohol) MQL oil 3 (based on mineral oil) MQL oil 4 (based on hydrocracked oil) MQL oil 5 (based on ester) MQL oil 6 (based on white oil) MQL oil 7 (based on carbon hydride) MQL water-miscible 1 (based on high polymer proportion) MQL water soluble 2 (based on synthetic oil) MQL water miscible 3 (based on mineral oil) Pure water

46 18 10 9.1 9 7.5 30 5% concentration 5% concentration 5% concentration ——

Table 1 Grinding conditions. Grinding mode Grinding wheel

Grinding machine Wheel speed (Vc) Work speed (Vft) Depth of cut (ae) Environments Conventional fluid grinding MQL oil flow rate Air pressure (also for air jet grinding) MQL nozzle distance from contact zone (d) and horizontal angle to the workpiece (a) Workpiece material Dresser Total depth of dressing (ad) Dressing speed (Vd) and overlap (Ud)

Plunge surface grinding, down cut 89A60I6V217 (vitrified bond corundum) 454A60I6V3 (vitrified bond SG) 89A60E6B22 (resin bond corundum) ELB micro-cut AC8 CNC Vc ¼30 m/s Vft ¼ 3000 mm/min ae ¼ 30 mm dry, fluid, air jet, MQL Syntilo XPS Castrol in a 5% concentration Q¼ 100 ml/hr P¼ 4 bar d¼ 80 mm, a ¼ 151 Hardened (100Cr6) with 50 72 HRC (60 mm  8 mm  13.8 mm) Two point diamond dresser ad ¼ 40 mm Vd ¼ 350 mm/min, Ud ¼ 2

ARTICLE IN PRESS 700

T. Tawakoli et al. / International Journal of Machine Tools & Manufacture 50 (2010) 698–708

Fig. 1. (a) Abrasive grains wear mechanisms of vitrified bond corundum after grinding (magnification: 175  ), (b) schematic of oil mist spray in MQL grinding and (c) lubricant sources at the interface of the grain and workpiece surface.

source and Vc represents the grinding wheel speed. Because the penetration depth of a gas strongly depends on its partial pressure, efficient transportation of oil droplets to the lubricant source needs to increase the oil mist mass flow rate and pressure [14]. Also, smaller droplets and low viscosity oils can effectively penetrate into the lubricant sources. It must be considered that the filling up time of oil mist (deviation of the lubricant sources passing time and penetration time, which can be explained as tfill ¼ (tpass tpen) ¼ (ls/Vc ls/Voilmist)) depends on the wheel speed. Consequently, by increasing the wheel speed, the filling up time of the oil mist will be decreased that leads to insufficient lubricating action. By increasing wear flats on the grains (dulling of the grits), the dimensions of the fractured grooves and therefore the

lubrication of the grain-workpiece contact zone will be reduced. During dressing of conventional wheels with a single-point diamond tool, the dresser follows a path which would appear to be like thread (fractured grooves) on the abrasive grains [7]. Coarser dressing causes more grain fracture and sharp wheel and consequently less contact area between the grain and workpiece in the grinding zone. Therefore, dressing process has significant effect on the MQL oil mist performance in the grinding process that must be considered to optimize MQL grinding. In addition, using grinding wheel with high porosity will result in more wheel pores and consequently more lubricant sources in the contact zone. Moreover, the wear mechanism of the SG abrasive that is characterized by grit micro fracture, favors a permanent self sharpening of the abrasive grains and prevents their flattening [15].

ARTICLE IN PRESS T. Tawakoli et al. / International Journal of Machine Tools & Manufacture 50 (2010) 698–708

Grinding wheel: 89A60I6V217(vitrified bond corundum) Material: 100Cr6 hardened steel Grinding parameters: ae= 30 µm, Vft=3000 mm/min, Vc=30 m/s, pass=300 MQL grinding with: Q=100 ml/hr, P=4 bar, d=80 mm

Tangential force, Ft (N)

Tangential force, Ft (N)

Grinding wheel: 89A60I6V217(vitrified bond corundum) Material: 100Cr6 hardened steel Grinding parameters: ae= 30 µm, Vft=3000 mm/min, Vc=30 m/s, pass=300 MQL grinding with: Q=100 ml/hr, P=4 bar, d=80 mm

701

Passes

Passes

Grinding wheel: 89A60I6V217(vitrified bond corundum) Material: 100Cr6 hardened steel Grinding parameters: ae= 30 µm, Vft=3000 mm/min, Vc=30 m/s, pass=300 MQL grinding with: Q=100 ml/hr, P=4 bar, d=80 mm

No orma al forc ce, Fn F (N N)

Normal force, Fn (N)

Grinding wheel: 89A60I6V217(vitrified bond corundum) Material: 100Cr6 hardened steel Grinding parameters: ae= 30 µm, Vft=3000 mm/min, Vc=30 m/s, pass=300 MQL grinding with: Q=100 ml/hr, P=4 bar, d=80 mm

Passes

Passes

Fig. 2. Grinding forces using vitrified bond corundum with: (a, c) MQL oil, (b, d) MQL water miscible oil, fluid, dry and air jet grinding.

Grinding forces can be separated into two parts, cutting deformation force and frictional force or sliding force (tangential sliding force; Ftsliding and normal sliding force; Fnsliding). The cutting deformation force is again sub-divided into ploughing force (tangential ploughing force; Ftploughing and normal ploughing force; Fnploughing) and cutting or chip formation force (tangential cutting force; Ftcutting and normal cutting force; Fncutting) [7]. Ploughing energy is expended by deformation of the workpiece material without removal. Ploughing is usually associated with side flow of material from cutting path into ridges, but it can also include plastic deformation of the material passing under the cutting edge [7]. On the conditions of lubrication action, the friction force in the grain wear flat area can be explained as Ff ¼ Ftsliding þFtploughing ¼ ts As þ tl Al where Ff is the friction force between the grain wear flat area and the workpiece surface, ts indicates the shearing strength of the workpiece material, As defines the adhesion area between the grain wear flat face and workpiece surface, tl represents the shearing strength of the boundary lubrication layer, Al the action area of the grain fracture and A¼As + Al explains the contact area between the grain flat face and workpiece. Another feasible and physics-based explanation that can consider the MQL grinding process is the so-called embrittlement action of the cutting fluid, which reduces the strain of facture of the work material [8,9]. This action is based on the Rebinder effect [8,9]. The results of Rebinder’s study showed that during cutting process absorbed fluid films prevent microcracks from closing (healing due to plastic deformation of the work material).

Because each microcrack in the machining zone serves as a stress concentrator, a lower energy was required for cutting (reducing the chip formation and ploughing energy components) [8,9]. Therefore, alternative feasible way for MQL to grinding process is to increase the embrittlement of the layer being removed and thus to reduce the work of plastic deformation done in the transformation of the layer being removed into the chip [8,9]. Consequently, during dry grinding process, cutting forces (Ftcutting) are higher than those in fluid and MQL grinding processes. It is pointed that with the rising contact zone temperatures, the grain cutting depth increases, since the material becomes more ductile because of the higher temperature [13], and consequently increases friction forces [16]. It must be noted that the formation of lubricating films, by chemical or by physical action can reduce workpiece-metal adhesion (smaller adhesion area between the grain wear flat and workpiece surface; As) and inhibit those chemical reactions which promote attritious wear [7]. The least wear flat area by far will be obtained with the grinding oils, and this leads to much lower sliding forces as well [7]. Accordingly, when the contact area between the grain flat surface and the workpiece (As) is decreased or the lubrication layer shearing strength (tl) is decreased, the friction force will be reduced. With high oil mist flow rate (Q) and high air pressure (P), the penetration performance of the grinding oil is much better and the friction force in the grain-workpiece interface will be reduced much more. Therefore, with oil mist as the coolant– lubricant better lubricating action can be obtained because of its excellent penetration performance and formation of low shearing strength lubrication layer [17].

ARTICLE IN PRESS 702

T. Tawakoli et al. / International Journal of Machine Tools & Manufacture 50 (2010) 698–708

Grinding wheel: 89A60E6B22 (resin bond corundum) Material: 100Cr6 hardened steel Grinding parameters: ae= 30 µm, Vft=3000 mm/min, Vc=30 m/s, pass=300 MQL grinding with: Q=100 ml/hr, P=4 bar, d=80 mm

Tangential force, Ft (N)

Tangential force, Ft (N)

Grinding wheel: 89A60E6B22(resin bond corundum) Material: 100Cr6 hardened steel Grinding parameters: ae= 30 µm, Vft=3000 mm/min, Vc=30 m/s, pass=300 MQL grinding with: Q=100 ml/hr, P=4 bar, d=80 mm

Passes

Passes Grinding wheel: 89A60E6B22 (resin bond corundum) Material: 100Cr6 hardened steel Grinding parameters: ae= 30 µm, Vft=3000 mm/min, Vc=30 m/s, pass=300 MQL grinding with: Q=100 ml/hr, P=4 bar, d=80 mm

Normal force, Fn (N)

Normal force, Fn (N)

Grinding wheel: 89A60E6B22 (resin bond corundum) Material: 100Cr6 hardened steel Grinding parameters: ae= 30 µm, Vft=3000 mm/min, Vc=30 m/s, pass=300 MQL grinding with: Q=100 ml/hr, P=4 bar, d=80 mm

Passes

Passes

Fig. 3. Grinding forces using resin bond corundum with: (a, c) MQL oil, (b, d) MQL water miscible oil, fluid, dry and air jet grinding.

4. Results and discussion The tangential and normal grinding forces versus passes and the surface roughness after 300th pass for all the grinding wheels and coolant–lubricants are shown in Figs. 2–5. The plotted forces are the average value of every 6 grinding passes. Among the base oils, hydrocracked oil seemed to generate lowest tangential grinding force while the other oils performed likely, although white oil (MQL oil 6) has slightly lower tangential grinding forces than ester and mineral oils at the beginning. Hydrocrack oils are mineral oil raffinates refined with hydrogen. Compared with mineral oils, they have a smaller aromate content, higher oxidation stability and higher viscosity [13]. Moreover, by means of a more homogeneous distribution of molecule size, they are clearly less prone to vaporization than mineral oils [13]. It is apparent that dry grinding with resin bond corundum has the highest grinding forces, while grinding with MQL oil and SG wheel has the lowest among the other abrasive and coolant–lubricant types. In the beginning (before 30 passes), all the forces for three abrasive wheels are comparable. With increasing passes, MQL grinding using different fluids exhibit different performance. Also for each grinding wheel while using grinding oils, normal grinding forces are approximately similar to those using water miscible oils. For vitrified bond SG wheel with all coolant–lubricant types, the grinding forces increase progressively which is not similar to the other abrasive types used in this study. It is clear that the coolant–lubricant in the grinding process influences the chip formation process by building up a lubricant

film, thus lowering the friction forces, and cooling the contact zone. As the lubrication effect increases, there is a corresponding increase in elastic–plastic deformation under the cutting edge of the abrasive grain, resulting in a decrease in workpiece roughness [18]. By reducing friction forces, friction heat and therefore the total process heat are reduced. However, too much lubrication (higher viscosity of oil) can have higher grinding forces, as the efficiency of the cutting process is reduced and relatively more energy is used in the shearing and deformation processes [18]. Also, higher viscosity oils adhere more strongly and produce less oil mist. The mechanism of the abrasive grain microwear is based on different physical–mechanical phenomena, like mechanical friction, adhesion, corrosion, diffusion and heat stress. On the other hand, abrasive grain fractures as well as bonding fracture caused by thermal, mechanical and chemical actions are the most important reasons of macrowear [3]. The lower grinding forces and friction generated when using the oil in comparison with when applying water miscible cause lower mechanical loads on the bond material and abrasive grains and consequently lower wheel wear. Fig. 5 indicates that the comparatively best roughness values were measured after grinding with vitrified bond corundum. The better surface values employing vitrified bond corundum can be explained by dulling of the grits. The dull grits and a film of MQL oil between the grit wear flat area and workpiece surface smooth the surface on the one hand and enlarge the deformation zone in the contact area on the other hand [19]. Therefore, with constant uncut chip thickness, the effective chip thickness (thickness of the

ARTICLE IN PRESS

Grinding wheel: 454A60I6V3 (SG) Material: 100Cr6 hardened steel Grinding parameters:ae= 30 µm, Vft=3000 mm/min, Vc=30 m/s, pass=300 MQL grinding with: Q=100 ml/hr, P=4 bar, d=80 mm

Grinding wheel: 454A60I6V3 (SG) Material: 100Cr6 hardened steel Grinding parameters: ae= 30 µm, Vft=3000 mm/min, Vc=30 m/s, pass=300 MQL grindingwith: Q=100 ml/hr, P=4 bar, d=80 mm

T nge Tan enttial fo orc ce, Ftt (N N)

Tangential force, Ft (N)

T. Tawakoli et al. / International Journal of Machine Tools & Manufacture 50 (2010) 698–708

passes

Grinding wheel: 454A60I6V3 (SG) Material: 100Cr6 hardened steel Grinding parameters: ae= 30 µm, Vft=3000 mm/min, Vc=30 m/s, pass=300 MQL grinding with: Q=100 ml/hr, P=4 bar, d=80 mm

Grinding wheel: 454A60I6V3 (SG) Material: 100Cr6 hardened steel Grinding parameters: ae= 30 µm, Vft=3000 mm/min, Vc=30 m/s, pass=300 MQL grinding with: Q=100 ml/hr, P=4 bar, d=80 mm

Normal force, Fn (N)

Normal force, Fn (N)

Passes

Passes

703

Passes

Fig. 4. Grinding forces using vitrified bond SG with: (a, c) MQL oil, (b, d) MQL water miscible oil, fluid, dry and air jet grinding.

formed chip) decreases simultaneously with a reduction of friction that results in lower surface roughness [10]. However, abrasive grain microfracture in SG wheels generates new cutting edges that protrude randomly from the surface of the grinding wheel and hence increases the surface roughness [15,16]. This wear mechanism favors the self sharpening of the grinding wheel. Concerning the grinding forces, this self sharpening phenomenon usually keeps the grinding forces within a limited range of variation in different coolant–lubricant conditions (Fig. 4). This would improve the work grindability and limit the surface damages [15]. These explanations are well illustrated by graphs of Figs. 2–4, that show higher levels of grinding forces when vitrified and resin bond corundum wheels are used. In addition, using SG wheel reduces grinding forces to two third and one third in comparison to those while grinding with vitrified and resin bond corundum wheels, respectively. Also, the process is less sensitive to the coolant–lubricant type when grinding with vitrified bond SG.

However, compared with vitrified bond corundum, resin bond corundum gives much higher forces. The possible explanation is that there is no interlocked structure with bond bridges (because there is minimal porosity). In the plots of grinding forces for all coolant–lubricants, it can be seen that the value of the force increases with increased number of passes except for resin bond corundum. Increasing of the grinding forces for vitrified bond corundum and SG wheels is due to the increased wheel wear with increased number of grinding passes. Also, for vitrified bond corundum at higher passes, the grinding forces remain flat after increasing in the beginning. This may be due to the hard and dense slurry layer observed on the grinding wheel. For resin bond corundum at higher passes, the grinding forces remain flat after decreasing in the beginning. This variation in data for resin bond corundum was caused by lower hardness of the resin bond corundum and higher fracture of the bond (high grinding forces on the grain can cause the bond posts to fracture

704

dry grinding

air jet grinding

air jet grinding

MQL oil 1

MQL oil 1

dry grinding air jet grinding MQL oil 1 MQL oil 2 MQL oil 3

MQL oil 3

MQL oil 3

MQL oil 4

MQL oil 4

MQL oil 5

MQL oil 5

MQL oil 6

MQL oil 6

MQL oil 7

MQL oil 7

MQL water miscible 1

MQL water miscible 1

MQL water miscible 2

MQL water miscible 2

MQL water miscible 3

MQL water miscible 3

MQL water miscible 3

MQL pure water

MQL pure water

MQL pure water

MQL oil 4 MQL oil 5 MQL oil 6 MQL oil 7 MQL water miscible 1 MQL water miscible 2

MQL Q

MQL

MQL oil 2

ARTICLE IN PRESS

dry grinding

T. Tawakoli et al. / International Journal of Machine Tools & Manufacture 50 (2010) 698–708

wet grinding wet grinding

MQL oil 2

Surface roughness (µm)

Surface roughness (µm)

wet grinding

MQL

Fig. 5. Surface roughness across the grinding direction for the ground specimens after 300th pass using: (a) vitrified bond corundum, (b) resin bond corundum and (c) vitrified bond SG.

Surface roughness (µm)

ARTICLE IN PRESS T. Tawakoli et al. / International Journal of Machine Tools & Manufacture 50 (2010) 698–708

705

Fig. 6. Abrasive grains loading for MQL grinding with MQL oil 1 after 300th pass (magnification: 100  ): (a, b) vitrified bond corundum before and after grinding, (c, d) resin bond corundum before and after grinding, (e, f) vitrified bond SG before and after grinding.

and the whole grain to be pulled out of the surface). In addition the high surface roughness for the resin bond corundum seems to be from the wheel loading and thicker slurry layer on the wheel. It is indicated in Figs. 2–4 that, grinding with fluid technique and MQL water miscible oils show significant improvement in grinding performance than dry, air jet and MQL pure water conditions by lowering the grinding forces. Using MQL fluid similar to that used in fluid grinding (MQL water miscible 1) showed nearly the same performances to conventional fluid grinding. These similar performances can be attributed to the enhanced penetration of the MQL oil mist, with low flow rate (100 ml/h) and its lubrication at the grinding interface. These results can be seen in MQL grinding using water miscible 3 (based on mineral oil). This shows that using water miscible oil with a small quantity results in the similar grinding performance in comparison to conventional fluid grinding process. This clearly demonstrates that, MQL system was able to penetrate into the region of contact between the grinding wheel and the workpiece more effectively than fluid cooling. Notice that the forces in MQL technique using pure water and vitrified bond corundum and also air jet grinding applying resin bond corundum

are increased exponentially after 144 passes and 78 passes, respectively. These large forces generate excessive heat that leads to visible burning of the workpiece, which can be identified by discoloration on the ground workpiece surface. Additionally, due to higher wear of the SG wheel, the surface roughness values increased for MQL pure water, air jet and dry grinding processes. The limitations of the MQL grinding technique with vitrified and resin bond corundum were caused by the wheel loading inducing an ascent of the process (Fig. 6). Vitrified bond SG wheel shows a different behavior in which wheel loading was a minor problem for MQL grinding using this abrasive type. The obvious differences in the morphology of surfaces ground with different abrasive and coolant–lubricant types suggest a considerable influence of the MQL grinding on the chip-formation mechanisms (Fig. 7). With regard to the surfaces of the ground specimens, it can be observed that there are hardly any side flows on surfaces ground with applying cutting oils in minimum quantity lubrication technique. However, surfaces generated with vitrified and resin bond corundum under dry, air jet and MQL with water miscible 1 conditions are characterized by

ARTICLE IN PRESS 706

T. Tawakoli et al. / International Journal of Machine Tools & Manufacture 50 (2010) 698–708

Fig. 7. Surface morphology of ground workpieces after 300th pass (magnification: 1000  ).

surface burning and more surface damages. It can be seen that in these conditions, lower thermal damages and material side flow can be observed when using SG grinding wheel. The chips resulted from grinding processes with different abrasive and coolant–lubricant types are illustrated in Fig. 8. It can be seen that the chips in dry, air jet and MQL grinding with water miscible oil conditions are much smaller, irregular and spherical. Chips produced under fluid condition are mainly long, thin and lamellar, indicating the mechanism of chip formation to be predominantly by low temperature. Minimum quantity lubrication technique with cutting oils also provided almost all types of chips indicating the mechanism of chip formation to be by higher temperature than fluid grinding. By vitrified bond corundum and SG grains, MQL grinding yielded almost similar types of chips suggesting approximately similar mechanism of chip formation with low temperature. Under MQL grinding with resin bond corundum the chips are welded to each other due to higher grinding forces and temperatures.

5. Conclusions The main results obtained in this study are summarized as follows:

1. It is apparent that MQL grinding with resin bond corundum has the least grinding performance, while the SG wheel shows an enhanced performance when using with grinding oils in this process. 2. The process is less sensitive to the coolant–lubricant type when grinding with vitrified bond SG. 3. SG wheels have the least tendency to the chip loading rather than vitrified and resin bond wheels. Moreover, the resin bond wheels are more prone to chip loading than vitrified bond wheels. 4. It was indicated that the lowest thermal damages and material side flow on the surface were generated by using the SG grinding wheel during MQL grinding process.

ARTICLE IN PRESS T. Tawakoli et al. / International Journal of Machine Tools & Manufacture 50 (2010) 698–708

707

Fig. 8. The chips obtained from different coolant–lubricant and abrasive types after 30th pass (magnification: 200  ).

5. Based upon the presented results, it was found that vitrified bond SG wheels and grinding oils have potential for the development of the MQL grinding process in comparison to vitrified and resin bond corundums and water miscible oils.

Acknowledgments The authors are indebted to TYROLIT Company for the supply of grinding wheels used in this research, and to Chemische Werke Kluthe GmbH (Germany) for its donation of the HAKUFORM coolant–lubricants used in the experiments. References [1] I.D. Marinescu, W.B. Rowe, B. Dimitrov, I. Inasaki, in: Tribology of Abrasive Machining Processes, William Andrew, USA, 2004.

[2] T. Tawakoli, E. Westkaemper, M. Rabiey, Dry grinding by special conditioning, International Journal of Advanced Manufacturing Technology 33 (2007) 419–424. ¨ [3] T. Tawakoli, E. Westkamper, M. Rabiey, A. Rasifard, Influence of the type of coolant–lubricant in grinding with CBN tools, International Journal of Machine Tools and Manufacture 45 (2007) 734–739. [4] T. Tawakoli, B. Azarhoushang, Influence of ultrasonic vibrations on dry grinding of soft stell, International Journal of Machine Tools and Manufacture 48 (14) (2008) 1585–1591. [5] T. Tawakoli, B. Azarhoushang, M. Rabiey, Ultrasonic assisted dry grinding of 42CrMo4, International Journal of Advanced Manufacturing Technology 42 (2009) 883–891. [6] T. Tawakoli, High efficiency deep grinding, VDI-Verlag and Mechanical Engineering Publications Limited (MEP) (1993). [7] S. Malkin, in: Grinding Technology: Theory and applications of machining with abrasives, 2nd ed., Industrial Press Inc., 2008. [8] J. Paulo Davim, in: Machining: Fundamentals and Recent Advances, Chapter 7- V.P. Astakhov; Ecological Machining: Near-dry Machining, Springer-Verlag London Limited, 2008. [9] V.P. Astakhov, in: Tribology of Metal Cutting, Elsevier Ltd., Great Britain, 2006. ¨ [10] T. Tawakoli, M.J. Hadad, M.H. Sadeghi, A. Daneshi, S. Stockert A. Rasifard, An experimental investigation of the effects of workpiece and grinding parameters on minimum quantity lubrication-MQL grinding, International Journal of Machine Tools and Manufacture 49 (2009) 924–932.

ARTICLE IN PRESS 708

T. Tawakoli et al. / International Journal of Machine Tools & Manufacture 50 (2010) 698–708

[11] G. Brunner, Schleifen mit mikrokristallinem Aluminiumoxid. Dr.-lng. ¨ Hannover, Fortschr.-Ber. VDI Reihe 2, no. 464, Dusseldorf, ¨ Diss., Universitat VDI-Verlag, 1998. [12] E. Brinksmeier, T. Brockhoff, A. Walter, in: Minimalmengenschmierung und Trockenbearbeitung beim Schleifen, HTM 52/3:166-170, Carl Hanser Verlag, ¨ Munchen, 1997. ¨ [13] Eh Fritz Klocke, Manufacturing process 2: grinding, honning, lapping, RWTH ed., ISBN: 978-3-540-92258-2, 2009. [14] T. Obikawa, Y. Asano, Y. Kamata, Computer fluid dynamics analysis for efficient spraying of oil mist in finish-turning of inconel 718, International Journal of Machine Tools and Manufacture 49 (2009) 971–978. [15] B.B. Fathallah, N.B. Fredj, H. Sidhom, C. Braham, Y. Ichida, Effects of abrasive type cooling mode and peripheral grinding wheel speed on the AISI D2 steel

[16] [17]

[18]

[19]

ground surface integrity, International Journal of Machine Tools and Manufacture 49 (2009) 261–272. I.D. Marinescu, M. Hitchiner, E. Uhlmann, W.B. Rowe, I. Inasaki, in: Handbook of Machining with Grinding Wheels, CRC Press, USA, 2007. J. Liu, R. Han, Y. Sun, Research on experiments and action mechanism with water vapor as coolant and lubricant in green cutting, International Journal of Machine Tools and Manufacture 45 (2005) 687–694. E. Brinksmeier, C. Heinzel, M. Wittmann, Friction cooling and lubrication in grinding, CIRP Annals, Manufacturing Technology 48 (2) (1999) 581–598. ¨ H.K. Tonshoff, H.G. Wobker, G. Brunner, F. Kroos, Limitations for lubricationfree grinding of hardened steel, Advancement of Intelligent Production (1994) 503–508.