High-power diode laser assisted hard turning of AISI D2 tool steel

High-power diode laser assisted hard turning of AISI D2 tool steel

ARTICLE IN PRESS International Journal of Machine Tools & Manufacture 46 (2006) 2009–2016 www.elsevier.com/locate/ijmactool High-power diode laser a...

813KB Sizes 0 Downloads 3 Views

ARTICLE IN PRESS

International Journal of Machine Tools & Manufacture 46 (2006) 2009–2016 www.elsevier.com/locate/ijmactool

High-power diode laser assisted hard turning of AISI D2 tool steel P. Dumitrescu, P. Koshy, J. Stenekes, M.A. Elbestawi McMaster Manufacturing Research Institute, and Department of Mechanical Engineering, McMaster University, Hamilton, Canada Received 14 December 2005; accepted 3 January 2006 Available online 28 February 2006

Abstract Difficulties with the integration of CO2 and Nd-YAG lasers into a machining centre have hitherto impeded industrial implementation of laser assisted machining technologies. The rapidly emerging new technology of high-power diode lasers holds potential in this regard, on account of the laser being compact and the feasibility of fiber optic beam transport. In comparison to other commercial systems, highpower diode lasers combine higher efficiency and metal absorption with lower capital and operating costs; however, their current metal processing applications are confined largely to surface hardening and joining, due to their lower power density. With a view to expanding their application envelope, the work presented in this paper explores high-power diode laser assisted turning of fully hardened AISI D2 tool steel, a material that is difficult to machine. Laser assist is shown to inhibit saw tooth chip formation, suppress chatter, deter catastrophic tool fracture and bring about a substantial reduction in tool wear and cutting forces, with minimal affect on the integrity of the machined surface. r 2006 Elsevier Ltd. All rights reserved. Keywords: Hard cutting; Laser assisted machining; Machinability

1. Introduction Enhanced flexibility, higher material removal rates, reduced lead times associated with better logistics, and environmental benefits of dry cutting have accelerated the widespread industrial adoption of hard part machining technology in the last decade. Hard cutting of AISI D2, which is a high-carbon, high-chromium tool steel widely employed in the manufacture of cold-forming tools for its excellent wear resistance and hardenability, presents major problems with respect to the current state of machining technology. A perspective of its extremely poor machinability can be realized with reference to typical tool life obtained when hard machining this material being an order of magnitude lower than that corresponding to AISI H13 tool steel [1]. In this context, this paper presents an enabling technology for hard turning D2 tool steel. Considering that machining by cutting entails shearing of the work material, one plausible avenue to improving the machinability of a difficult-to-cut material is to reduce Corresponding author. Tel.: +1 905 525 9140; fax: +1 905 572 7944.

E-mail address: [email protected] (P. Koshy). 0890-6955/$ - see front matter r 2006 Elsevier Ltd. All rights reserved. doi:10.1016/j.ijmachtools.2006.01.005

its shear strength selectively in the immediate vicinity of the shear zone by thermal softening. This technique known as hot machining could be accomplished through the use of an external heat source such as a laser, the role of which in this instance is to merely assist the cutting process, but not to be directly engaged in material removal as in laser machining. The work reported herein explores laser assisted turning of hardened D2 tool steel. The novel aspect of this work is the application of a High-Power Diode Laser (HPDL), the technological characteristics of which are quite different from CO2 and Nd-YAG lasers hitherto used elsewhere for Laser Assisted Machining (LAM). 2. LAM: A brief overview The technology of hot machining is not new: the first patent on hot machining was issued before the introduction of high speed steel. Previous generations of this technology employed low-grade heat sources such as flame, electrical resistance, induction and plasma arcs. With the advent of several advanced difficult-to-cut materials, and with the availability of heat resistant tool materials and

ARTICLE IN PRESS 2010

P. Dumitrescu et al. / International Journal of Machine Tools & Manufacture 46 (2006) 2009–2016

cost-effective lasers, there has been a renewed interest in this technology. Successful application of LAM relates to the following advantages: increase in material removal rates, reduction in the occurrence of chatter and catastrophic tool failure, decrease in cutting forces and tool wear, and the capability to cut brittle work materials without extensive cracking [2]. In profile machining applications, the ability to cut rather than grind a material is further advantageous with regard to process flexibility. The challenge in realizing an ideal LAM process is to ensure that the rate of heating is commensurate with the kinematic cutting parameters such that the bulk of the input heat is transported in the chip, with the least thermal detriment to the machined surface/sub-surface. Work material classes that are well suited to the application of LAM technology include: (i) hard and brittle materials such as engineering ceramics that can otherwise be subject only to cost-prohibitive abrasive processing, (ii) heat resistant materials like nickel alloys, (iii) materials with abrasive constituents such as high silicon content aluminum alloys, and (iv) materials with a propensity to significant strain hardening like austenitic stainless steels. Klocke and Zaboklicki [3] presented several demonstrator components that confirmed the feasibility of cutting sintered silicon nitride through LAM. For workpiece temperatures in the range of 1000–1400  C, chip formation was observed to be realized predominantly through ductile plastic deformation, resulting in machined surfaces of a quality similar to that obtained in diamond grinding. Based on electron microscopic analysis of chips obtained, Lei et al. [4] inferred that plastic deformation of silicon nitride in the shear zone was sustained by the enhanced mobility of the rod-like silicon nitride grains, facilitated by a reduction in the viscosity of the intergranular glassy phase at elevated workpiece temperatures. Rebro et al. [5] developed a double-ramp laser profile protocol for LAM of a mullite ceramic to preclude thermal fracture of the workpiece, the incidence of which is exacerbated by the lower thermal diffusivity, fracture toughness and tensile strength of this porous material, in comparison to silicon nitride. LAM was further employed by Pfefferkorn et al. [6] to cut a partially opaque zirconia, and achieved a forty-fold improvement in the life of the polycrystalline cubic boron nitride tools (PCBN) in an operating workpiece temperature window of 900–1100  C. LAM of an alumina reinforced aluminum metal matrix composite was studied by Wang et al. [7] who reported that laser assist reduced cutting forces and wear of the carbide tool by 30–50% and 20–30%, respectively. The machined surface was further observed to exhibit enhanced wear resistance, due to the higher concentration of alumina particles in and immediately beneath the generated surface, evidently due to the tool nose displacing the particles into the thermally softened aluminum matrix. Gratias et al. [8] characterized the effect of laser power and beam-tool lead distance in the machining of hardened XC42 steel (equivalent to AISI 1042) and found the cutting forces to

be reduced by 80%. During the course of an investigation on LAM of aerospace alloys, Lesourd et al. [9] observed laser heating to inhibit catastrophic shear instability in the cutting of titanium alloys, and facilitate machining of Inconel at speeds as high as 400 m/min, which represents an order of magnitude productivity enhancement. LAM applications detailed above predominantly pertain to turning processes, although schemes have been outlined [10] for milling. Westkaemper [11] has further reported a laser assisted grinding process wherein the stock removal rate in continuous dress grinding of hot pressed silicon nitride was increased by a factor of six with no adverse effects on the machined surface quality. Despite having been introduced more than two decades ago and the large body of research literature that has accrued over this period, there seems to be no large scale industrial implementation of LAM technology at present. This is predominantly due to issues referring to safe and seamless integration of the laser into the machine tool. The relatively new and emerging technology of HPDL [12,13] holds significant potential in this regard. 3. HPDL in the context of LAM Individual diode lasers are a few hundred micrometers in size, and have a limitation on their output power. Depending on the total power and beam quality required, HPDL therefore constitute several diode laser stacks comprising a number of diodes. Although the output is therefore multimodal, and incoherent, the beam is of high temporal stability, which is critical for the robustness of a manufacturing process. The output radiation of the diode diverges 2–6 times higher in a plane perpendicular to the plane of the p–n junction known as the fast axis, in comparison to the parallel plane identified as the slow axis. Due to fast and slow axis collimations, the laser beam footprint of commercial direct diode systems are generally rectangular, which influences the design and response of the process that the laser is used for. In reference to the application of HPDL for LAM, its characteristics vis-a`-vis CO2 and Nd-YAG lasers are summarized in Table 1. Compared to CO2 lasers, HPDL operate at a much lower wavelength band, resulting in better metal absorption and smaller absorption length, both of which have significant process implications. Application of an absorption enhancing coating is an option in laser processing, however, this represents an additional operation and is not quite practicable for multi-pass LAM processes. To this end, HPDL hold a definite advantage over CO2 lasers. The shorter absorption length is also beneficial for LAM applications in terms of minimizing thermal damage to the generated surface. HPDL further possess higher electrical to optical conversion efficiency, resulting in less severe cooling requirements, which reflects favorably on the size of the laser head and the peripheral equipment. Compactness is critical to integrating the laser into the machine tool; so is the feasibility of conveying the laser

ARTICLE IN PRESS P. Dumitrescu et al. / International Journal of Machine Tools & Manufacture 46 (2006) 2009–2016

2011

Table 1 A comparison of laser systems

Wavelength (mm) Metal absorption (%) Efficiency (%) Laser head size (cm3 =W) Maximum power (kW) Maximum power intensity (W=cm2 ) Capital cost ($/W)

CO2

Nd-YAG

HPDL

10.6 5–10 10–15 1000 40 108 150–300

1.06 25–35 3–5 10 4 109 200–600

0.8–1.0 30–40 30–50 1 10 106 100–300

beam using a fiber optic from outside of and within the confines of a machine tool, which is possible only with HPDL and Nd-YAG lasers. Capital and operating costs of HPDL are however significantly lower than Nd-YAG lasers, in addition to the former being also essentially maintenance-free. Despite the continuous and rapid progress in technology, current HPDL systems have limitations such as low power density, poor focusability, low working distances, and an expected operational lifetime of only about 10,000 h. Presently, direct HPDL of power 10 kW and spot size 0:6  3:0 mm2 , and fiber-coupled lasers of power 3 kW and 600 mm fiber diameter are commercially available. The bulk of current HPDL applications are limited to surface modification and joining technologies due to their lower power density [14]. Beam dimension need to be reduced to less than 200 mm for HPDL to be suitable for metal removal applications that involve melting and/or vaporization [12]. In the context above, it is interesting to note that LAM does indeed constitute an appropriate HPDL application, since in this case it is preferable that the work material be only thermally softened with no melting. A survey of the literature however indicates that there is only a single report of such an application. LAM of silicon nitride using HPDL is alluded to in [13], but there is neither comprehensive information nor experimental results presented. Barnes et al. [15] reported on the improvement in the machinability of a metal matrix composite pretreated using a HPDL, but since machining in this case follows pretreatment as a separate operation, this cannot be considered LAM. Work detailed in the present paper appears to be one of the first instances of accomplishing LAM using a HPDL. 4. Experimental The objective of the experimental work was to benchmark the effectiveness of HPDL assisted hard turning of AISI D2 tool steel of hardness 60 HRC in terms of tool life, cutting forces and workpiece surface integrity. LAM experiments reported in this paper involved a continuous wave direct HPDL system with a maximum rated power of 2 kW, a beam size of 4:0  0:9 mm2 , a working distance of 85 mm, and an operational wavelength range of

Fig. 1. Laser assisted machining set-up.

0.81–0:94 mm. The laser head dimensions of only 410  150  130 mm3 allowed simple integration of this system into a turning centre (Fig. 1) with several safety interlocks. The machine tool was further specially equipped to eliminate any diffuse laser scattering hazard, and be in compliance with all engineering controls prescribed by ANSI Z136.1 [16] to operate a fully enclosed Class IV laser as a Class I laser. With the aid of appropriate safety glasses, the process could be observed through a window in the turning centre, fitted with a laser grade filter glass of an optical density of approximately 6. The laser head was held in a fixture attached to the turret of the turning centre to position the laser spot ahead of the cutting tool edge, and was further inclined with respect to the machined surface so as to avoid the reflected beam from re-entering the optical system and causing damage. Compressed air at a pressure of 2 bar was used to protect the laser optics from the fumes and debris resulting from the heating and cutting processes. Workpieces were in the form of disks of diameter 100 mm and thickness 3.5 mm in order to minimize possible spatial variations in work material hardness, and were held in a mandrel supported between the chuck and the tailstock. With reference to the rectangular footprint of the direct HPDL, its efficacy in assisting hard turning D2 tool steel was first investigated in an orthogonal machining configuration simulating a parting-off operation, with the laser slow axis parallel to the work axis such that the entire cut width was simultaneously heated (Fig. 2a). This was followed by longitudinal turning experiments wherein the effect of orienting the slow axis along and across the feed direction (Figs. 2b and c) was studied. Due to constraints with the accommodation of the laser head and the

ARTICLE IN PRESS P. Dumitrescu et al. / International Journal of Machine Tools & Manufacture 46 (2006) 2009–2016

2012

work slow axis

Flank wear (µm)

180 140 conventional cutting 100 60 LAM

20

fast axis 0

(a) laser spot

100

150

200

250

300

(b)

(c)

Flank wear (µm)

400

tool (a)

50

300 conventional cutting

200

LAM (3 repeat trials)

100 0

Fig. 2. Plan view of laser assisted turning configurations investigated: (a) orthogonal cutting; (b) longitudinal turning with laser slow axis parallel to tool feed, and (c) longitudinal turning with laser slow axis perpendicular to tool feed.

measurement hardware in the working space of the lathe, the lead distance between the laser spot and the cutting edge was 25 mm. Orthogonal experiments entailed a 4 mm wide grooving tool that held a flat face TiN–coated carbide insert with rake and relief angles of 0 and 7 , respectively. Longitudinal turning involved a tool with a TiAlN-coated flat face carbide insert of 5 rake angle, 11 relief angle, and 45 side cutting edge angle. All LAM experiments were performed at a laser power of 2 kW. Orthogonal machining trials involved cutting speeds of 20, 30 and 40 m/min, a radial infeed of 0.05 mm/rev, and a cut width equal to the work thickness of 3.5 mm. Longitudinal turning experiments related to a cutting speed of 100 m/min, a feed of 0.05 mm/rev and a radial depth of cut of 0.5 mm. Conventional cutting experiments (with no laser assist) were also conducted for the purpose of comparison. All cutting tests were done dry. Cutting forces were measured using a three-axis dynamometer. Work surface temperature was recorded at a point 10 mm in front of the cutting edge using an infrared thermometer capable of measuring emissive surface radiation at temperatures exceeding 350  C. Flank wear was monitored using a tool maker’s microscope. Chip samples collected during the cutting tests, and sections of machined surfaces were mounted in an epoxy base, and then polished and etched with a 2% nital solution. Microhardness of the workpiece subsurface was characterized at a load of 1 kgf. 5. Results and discussion 5.1. Orthogonal cutting Figs. 3a and b depict the effect of laser assist on the evolution of flank wear in respect of conventional

0 (b)

20

40

60 80 100 Length cut (m)

120

140

160

Fig. 3. Comparison of tool flank wear in conventional and laser assisted orthogonal machining: (a) 20 m/min cutting speed, (b) 30 m/min cutting speed.

machining at cutting speeds of 20 and 30 m/min, respectively. The plots indicate the sensitivity of tool performance to cutting speed when cutting conventionally: while the wear rate was acceptable at a speed of 20 m/min, an incremental increase of 10 m/min was found to precipitate catastrophic tool fracture within the first few meters of length cut. To ensure that this was not an isolated occurrence, this experiment was repeated several times with consistent outcome. This confirmed that a cutting speed of 20 m/min represented the limit for this rather intricate process that engages the entire tool width on a hard material that is difficult to machine. The application of laser assist was found to enhance tool performance both in terms of reducing tool wear as well as averting catastrophic failure. The evolution of flank wear at a cutting speed of 20 m/min shown in Fig. 3a indicates that the application of laser assist corresponds to a significantly lower tool wear rate, which manifests a twofold reduction in flank wear at a length cut of 200 m. More remarkable is the enabling aspect of LAM at a cutting speed of 30 m/min (Fig. 3b). While mechanical cutting repeatedly resulted in catastrophic tool failure, test replications indicated the application of laser assist not only to facilitate cutting, but also render the wear process both progressive and repeatable. The modes of tool wear under conditions of conventional cutting comprised severe abrasive wear (Fig. 4a), and/or large scale chipping of the cutting edge (Fig. 4b) that eventually led to extensive thermal damage culminating in abrupt tool failure. Such wear modes are characteristic of hard cutting D2 tool steel, the microstructure of which typically constitutes a martensitic matrix embedding primary ðFe; CrÞ7 C3 carbides that are 5–20 mm in size.

ARTICLE IN PRESS P. Dumitrescu et al. / International Journal of Machine Tools & Manufacture 46 (2006) 2009–2016

1000 0 -1000

axial

3000 2000 1000

thrust

0 2000 1000 0

cutting 0

(a)

5

10

Time (s)

15

20

0

5

(b) Temp. (°C)

These blocky carbides that have been found to be responsible for the enhanced abrasive wear resistance of D2 tool steel [17], concomitantly have an adverse influence on its machinability. While characterizing the performance of PCBN tools in cutting hardened tool steels of identical hardness, Poulachon et al. [18] observed hardened D2 tool steel to induce particularly high wear rates, which was attributed to the massive carbides of hardness 2400 HV (roughly three times of that of martensite). This was evident from the width of grooves that comprise conspicuous features in the flank wear morphology (see Fig. 4a) corresponding to the size of the carbides or carbide clusters. Catastrophic tool fracture, when hard cutting this steel has further been reported in turning applications employing mixed alumina tools [19] and milling applications using carbide/PCBN tooling [20]. The extent of the difficulty in machining hardened D2 tool steel can be appreciated in reference to the work of Becze et al. [21] who performed high temperature mechanical testing of this material using a Compressive Split Hopkinson Bar incorporating a punching shear strain state, which involved strains and strain rates comparable to those in cutting. Owing to the secondary carbide phase presenting numerous pinning sites that impede the mobility of dislocations, the material was observed to strain harden significantly over the entire strain range tested, at both high and low strain rates. Accordingly, the shear strength at failure was measured to be in excess of 1300 MPa at ambient test temperature. The carbide phase thus hampers the machinability of hardened D2 both in terms of increasing the flow stress of the material and inflicting severe abrasive wear on the tool.

The incidence of a stable built-up edge shielding the cutting edge was observed during LAM, and is deemed to be responsible in part for the reduction in tool wear. More prominent is the benefit of work material thermal softening leading to a reduction in the flow stress brought about by laser heating that: (i) debilitates the abrasive carbides in the work material from inducing severe tool wear, and (ii) reduces the stresses acting on the tool to a level that precludes brittle fracture. The effect of material softening due to laser assist can be explicitly seen in the progression of cutting forces at a cutting speed of 20 m/min shown in Fig. 5. The conventional cutting process was extremely loud and exhibited severe chatter (Fig. 5a). Although the process was configured to be two-dimensional, axial forces were significant due to the lateral displacement of the tool linked to the high thrust forces. Application of laser assist (Fig. 5b) completely eliminated chatter, reinstated the process to be orthogonal, and led to a significant reduction in thrust force, which signifies thermal softening of the martensitic matrix, given that the alloy carbides are stable at high temperatures. At elevated temperatures, D2 has been reported [21] to exhibit increased ductility, with the failure mode changing from being predominantly fracture up to a test temperature of 650  C to plastic deformation at 750  C, which is roughly half the melting temperature (1490  C) of the alloy. Referring to the measured temperature in relation to the cutting forces (Fig. 5c), it appears that a surface temperature of 300–400  C is sufficient to induce appreciable softening of the work material across the uncut chip thickness of 0.05 mm. The work material transition temperature and the work surface temperature due to laser heating above seem to be largely in agreement, considering that under conditions of conventional machining, the shear zone temperature would be in excess of 300  C [22], which

Force components (N)

Fig. 4. Tool failure modes in conventional machining: (a) abrasive wear, and (b) edge chipping.

2013

15

20

15

20

600 500 400 300 0

(c)

10

Time (s)

5

10

Time (s)

Fig. 5. Force components in orthogonal cutting: (a) conventional machining, (b) LAM; (c) shows corresponding surface temperatures in LAM.

ARTICLE IN PRESS 2014

P. Dumitrescu et al. / International Journal of Machine Tools & Manufacture 46 (2006) 2009–2016

need to be taken into account in conjunction with the effect of external heating. It is interesting to note that even a modest temperature rise across the uncut chip thickness due to external heating would bring about a significant net temperature rise in the shear zone, since the associated increase in work material ductility would translate into additional heat generation linked to the more extensive plastic deformation. Furthermore, the temperature gradient across the uncut chip thickness due to laser heating can be expected to be compensated for by the cutting-induced high temperatures that pertain in the vicinity of the tool tip. With continued external heating leading to higher workpiece temperatures, the proportion of heat generated in the shear zone due to plastic deformation is however bound to decrease, due to the monotonic reduction in the shear flow stress of the material. Heat transfer modeling of the process incorporating high temperature work material behaviour is essential to formulate an appropriate theoretical framework for the design of an optimal process, which is currently underway and will be reported in due course. The thermal softening effect is conclusively demonstrated by the chip morphology changing from saw tooth form to continuous chip on application of laser assist (Fig. 6). The decrease in chip thickness when the cutting speed is increased from 20 to 30 m/min (Figs. 6b and c) when machining with laser assist refers to the decreased reduction in the hardness of the work material due to the less effective heating at the higher speed. LAM experiments at speeds higher than 40 m/min indicated the maximum power of 2 kW to be insufficient to obviate catastrophic fracture of the tool. In light of its lower power density, the maximum cutting speed that can be effectively realized in a turning process is limited by the laser power, as the rate of laser heating has to be consistent with the cutting speed. This is however not as crucial for a milling process, as the laser scanning speed need only relate to the feed speed, which is generally much lower than the cutting speed.

5.2. Longitudinal turning In contrast to the circular beam shape of a fiber coupled laser, direct HPDL have a rectangular footprint with the consequence that the orientation of the laser spot in respect of the process kinematic scheme would have an influence on the process response. This effect can be quite significant, depending on the aspect ratio of the beam. Pinkerton and Li [23] investigated the effect of beam orientation in relation to its traverse direction in a direct metal deposition application, and found the wall height obtained to be 22% higher when the beam moves perpendicular to the fast axis, for a beam aspect ratio of 1.4. In a surface hardening process employing a HPDL with a beam aspect ratio in the range of 2.0–2.8, Klocke et al. [24] report a 40% reduction in the volume of material hardened per unit time, when the feed direction is along rather than across the slow axis; a laser traverse along the slow axis was however found to be appropriate for welding. As discussed previously, the beam footprint of the HPDL used in this work was 0.9 mm along the fast axis and 4.0 mm along the slow axis. As both these dimensions are much higher than the feed per revolution typically used in a longitudinal turning process, the role of the laser assist was assessed with respect to the laser slow axis oriented along (Fig. 2b) and across (Fig. 2c) the feed direction, which represent bounds encompassing all possible orientations. In a work investigating hard turning of D2 tool steel using PCBN tools, Arsecularatne et al. [25] found speeds higher than 140 m/min to be not viable in the interest of tool life, and further noted the optimal cutting speed to be on the order of 70 m/min. Laser assisted longitudinal experiments reported in the present paper employed coated carbide tools and a cutting speed of 100 m/min, which therefore represents a significant advance in both productivity and tooling cost. When machining conventionally, tool life for a flank wear criterion of 300 mm corresponded to a length cut of 160 m (Fig. 7). The flank wear morphology pointed to severe abrasion and micro-chipping of the cutting edge as the wear modes, as reported also in [25]. Laser assist improved tool life roughly by 100%, with tool wear referring to laser slow axis oriented perpendicular to the

Flank wear (µm)

400 300

LAM (slow axis parallel to feed) conventional cutting

200 100 LAM (slow axis perpendicular to feed)

0 0

50

100

150

200

250

300

350

400

Length cut (m) Fig. 6. Representative chip forms obtained in: (a) conventional cutting at 20 m/min cutting speed; (b) LAM at 20 m/min cutting speed, and (c) LAM at 30 m/min cutting speed.

Fig. 7. Comparison of tool flank wear in conventional and laser assisted longitudinal turning.

ARTICLE IN PRESS P. Dumitrescu et al. / International Journal of Machine Tools & Manufacture 46 (2006) 2009–2016

follows each heating cycle, the role of which becomes more dominant with increasing workpiece diameter. The increase in workpiece temperature with time (Figs. 5c, 8b and c) due to the accumulation of the residual heat that conducts past the uncut chip cross-section into the workpiece bulk represents another limitation of HPDL. For applications such as hard cutting, this could lead to undesired overtempering of the workpiece resulting in a loss of hardness. This problem could be addressed by cooling the workpiece through external means or by continuous control of incident laser power with feedback based on surface temperature or cutting forces. Measurement of hardness (Fig. 9) however indicated no significant softening in the machined surface and subsurface, which is to be expected as D2 tool steel is highly resistant to back tempering, with only a minimal loss in hardness for temperatures up to 550  C [26]. If thermal damage ensues and is of concern, finish machining could be accomplished with no laser assist so as to remove the heat affected layer.

1000 Vickers hardness

feed direction being somewhat lower than that oriented parallel. Flank wear under conditions of LAM constituted structural features akin to conventional cutting, but on a significantly reduced scale. Similar to what was observed in orthogonal turning experiments (Fig. 6), laser assist was found to suppress saw tooth chip formation in longitudinal turning, but less so with the laser slow axis oriented parallel to the feed direction. The evolution of cutting and axial forces components when machining conventionally was not significantly different from that of LAM. The thrust force was however significantly reduced in LAM as expected, due to work material thermal softening, confirmed by surface temperature measurements (Fig. 8). The maximum surface temperature attained with the laser slow axis parallel to the feed direction was about 450  C, while that of the slow axis perpendicular to the feed direction was in excess of 600  C. The effect of the orientation of the laser spot on surface temperature above can be comprehended in reference to: (i) the number of heating/cooling cycles, (ii) the duration of heating, and (iii) the distribution of the laser power density along the fast and slow axes. Laser heating assisting machining in a turning process involves cyclic heating. For the rectangular beam of size 0:9  4:0 mm2 used in this work, orienting the slow axis parallel to the feed direction corresponds to about 4 times as many heating cycles as the slow axis oriented perpendicular. As the surface temperature was nevertheless lower in the former case, it is evident that the temperature is dictated rather by the duration of heating in a single cycle. This is further reinforced by the power density distribution being top-hat along the laser slow axis, as opposed to Gaussian across. The effect of the number of heating cycles can be rationalized in terms of the relatively longer interval associated with cooling that

2015

900

slow axis perpendicular to feed

800 700 600

slow axis parallel to feed bulk hardness

500 400 0.0

0.2 0.6 0.8 1.0 1.2 0.4 Distance from machined surface (mm)

1.4

Fig. 9. Microhardness profile of surfaces generated in laser assisted longitudinal turning, for different laser orientations.

Temp. (°C)

700

Thrust force (N)

500 400 300

500 400 300 200 100 0 (a)

600

0

2

4 Time (s)

6

8

0 (b)

2

4 Time (s)

6

8

0 (c)

2

4

6

8

Time (s)

Fig. 8. (a) Thrust force in conventional longitudinal cutting; (b) temperature profile and thrust force in LAM with laser slow axis oriented parallel to feed direction; (c) temperature profile and thrust force and in LAM with laser slow axis oriented perpendicular to feed direction.

ARTICLE IN PRESS 2016

P. Dumitrescu et al. / International Journal of Machine Tools & Manufacture 46 (2006) 2009–2016

6. Conclusions The paper demonstrated laser assisted machining of fully hardened AISI D2 tool steel using a high-power diode laser in orthogonal and longitudinal turning processes. In respect of a conventional process, laser assist was found to: (i) obviate catastrophic fracture of the carbide tools used, (ii) improve tool life by as much as 100%, (iii) suppress machining chatter and saw tooth chip formation, and (iv) reduce the thrust component of the cutting force. Furthermore, laser heating was of no significant thermal detriment to the machined surface. Orienting the laser slow axis perpendicular to the feed direction is more effective in assisting a longitudinal turning process, as the duration of heating rather than the number of heating cycles is the dominant factor that influences workpiece temperature. For a certain maximum laser power and work material, HPDL systems are effective only below a threshold cutting speed in a turning process, consequent to its power density limitation. They would however be better suited to assisting milling applications, wherein the rate of laser heating need correspond only to the feed speed, as opposed to the cutting speed in turning that is generally higher. In light of their ease of integration into a machine tool, higher efficiency, better metal absorption, and lower capital/operating costs in comparison to CO2 and NdYAG lasers, HPDL can be expected to pave the way for large scale industrial implementation of laser assisted machining technologies in the future. Acknowledgments This work was funded by: AUTO21 (www.auto21.ca), a national research initiative supported by the Government of Canada, and Natural Sciences and Engineering Research Council of Canada (www.nserc.ca). References [1] C.E. Becze, A thermo-mechanical force model for machining hardened steel, Doctoral Thesis, McMaster University, 2002. [2] G. Chryssolouris, N. Anifantis, S. Karagiannis, Laser assisted machining: an overview, Journal of Manufacturing Science and Engineering 119 (1997) 766–769. [3] F. Klocke, A. Zaboklicki, Laser assisted turning of ceramics, Machining of Ceramics and Composites, Marcel Dekker, New York, 1998, 551–574. [4] S. Lei, Y.C. Shin, F.P. Incropera, Deformation mechanisms and constitutive modeling of silicon nitride undergoing laser assisted machining, International Journal of Machine Tools and Manufacture 40 (2000) 2213–2233. [5] P.A. Rebro, Y.C. Shin, F.P. Incropera, Laser assisted machining of reaction sintered mullite ceramics, Journal of Manufacturing Science and Engineering 124 (2002) 875–885.

[6] F.E. Pfefferkorn, Y.C. Shin, Y. Tian, F.P. Incropera, Laser assisted machining of magnesia-partially-stabilized zirconia, Journal of Manufacturing Science and Engineering 126 (2004) 42–51. [7] Y. Wang, L.J. Yang, N.J. Wang, An investigation of laser assisted machining of Al2 O3 particle reinforced aluminum matrix composite, Journal of Materials Processing Technology 129 (2002) 268–272. [8] J.F. Gratias, L.J. Fan, G. Marot, P. Cohen, A. Moisan, Proposition of a method to optimize the machining of XC42 steel with laser assistance, CIRP Annals 42 (1993) 115–118. [9] B. Lesourd, T. Thomas, L’usinage assiste par laser, Me´canique Industrielle et Mate´riaux 48 (1995) 22–24. [10] H. Hugel, M. Wiedmaier, T. Rudlaff, Laser processing integrated into machine tools: design, applications and economy, Optical and Quantum Electronics 27 (1995) 1149–1164. [11] E. Westkaemper, Grinding assisted by Nd-YAG lasers, CIRP Annals 44 (1995) 317–320. [12] L. Li, The advances and characteristics of high-power diode laser materials processing, Optics and Lasers in Engineering 34 (2000) 231–253. [13] P. Loosen, H.G. Treusch, C.R. Hass, U. Gardenier, M. Weck, V. Sinnhoff, S. Kasperowski, R. Esche, High power diode lasers and their direct industrial applications, Proceedings of the International Society for Optical Engineering 2382 (1995) 78–88. [14] E. Kennedy, G. Byrne, D.N. Collins, A review of the use of high power diode lasers in surface hardening, Journal of Materials Processing Technology 155–156 (2004) 1855–1860. [15] S. Barnes, R. Morgan, A. Skeen, Effect of laser pre-treatment on the machining performance of aluminum/SiC MMC, Journal of Engineering Materials and Technology 125 (2003) 378–384. [16] American National Standard for Safe Use of Lasers, ANSI Z136.1, American National Standards Institute, Washington, DC, 2000. [17] L. Bourithis, G.D. Papadimitriou, J. Sideris, Comparison of wear properties of tool steels AISI D2 and O1 with the same hardness, Tribology International 39 (2006) 479–489. [18] G. Poulachon, S. Pheulpin, E. Seguin, B.P. Bandyopadhyay, I.S. Jawahir, The influence of the microstructure of hardened tool steel workpiece on the wear of PCBN cutting tools, International Journal of Machine Tools and Manufacture 43 (2003) 139–144. [19] J.G. Lima, R.F. Avila, A.M. Abrao, M. Faustino, J.P. Davim, Hard turning: AISI 4340 high strength low alloy steel and AISI D2 cold work tool steel, Journal of Materials Processing Technology 169 (2005) 388–395. [20] P. Koshy, R.C. Dewes, D.K. Aspinwall, High-speed end milling of hardened AISI D2 tool steel, Journal of Materials Processing Technology 127 (2002) 266–273. [21] C.E. Becze, M.J. Worswick, M.A. Elbestawi, High strain rate shear evaluation and characterization of AISI D2 tool steel in its hardened state, Machining Science and Technology 5 (2001) 131–149. [22] E. Ng, D.K. Aspinwall, The effect of workpiece hardness and cutting speed on the machinability of AISI H13 hot work die steel when using PCBN tooling, Journal of Manufacturing Science and Engineering 124 (2002) 588–594. [23] A.J. Pinkerton, L. Li, An experimental and numerical study of the influence of diode laser beam shape on thin wall direct metal deposition, Journal of Laser Applications 17 (2005) 47–56. [24] F. Klocke, A. Demmer, A. Zaboklicki, Investigation into the use of high power diode lasers for hardening and thermal conduction welding of metals, Proceedings of the International Society for Optical Engineering 3097 (1997) 592–599. [25] J.A. Arsecularatne, L.C. Zhang, C. Montross, P. Mathew, On machining of hardened AISI D2 steel with PCBN tools, Journal of Materials Processing Technology 171 (2006) 244–252. [26] ASM Handbook, vol. 4, Heat Treating, ASM International, Metals Park, Ohio, 1991.