j o u r n a l o f m a t e r i a l s p r o c e s s i n g t e c h n o l o g y 2 0 1 ( 2 0 0 8 ) 285290

journal homepage: www.elsevier.com/locate/jmatprotec

Laser cutting of thick sheet metals: Effects of cutting parameters on kerf size variations
B.S. Yilbas
ME Department, KFUPM, Dhahran, Saudi Arabia

a r t i c l e
Keywords: Laser Cutting Thick sheets Striation

i n f o

a b s t r a c t
Laser cutting of thick sheets, in general, results in poor end product quality due to excessive thermal erosion from the kerf sites. In the present study, laser melting of thick mild steel sheets is considered and effect of cutting parameters on the percentage of kerf width variation is examined. The cutting parameters considered include laser output power, cutting speed, and oxygen assisting gas pressure. A factorial analysis is carried out to identify the main effects and interactions of the parameters. Thermal efciency of cutting and liquid layer thickness is formulated. Optical microscopy and scanning electron microscopy (SEM) are carried out to examine the cutting defects and the kerf size variation. It is found that laser output power and oxygen gas pressure have signicant effect on the percentage of kerf width variation. 2007 Elsevier B.V. All rights reserved.

1.

Introduction

Laser cutting of thick sections offers considerable advantages over the conventional techniques due to precision of operation, short processing time, and low cost. The physical processes involved in laser cutting of thick sections are complicated and signicantly inuences the end product quality. Laser parameters, in particular laser output power, focus on setting of focusing lens, cutting speed, assisting gas, and its pressure inuence the physical processes in the cutting section. In this case, controlling the affecting parameters results in improved cutting quality. Consequently, investigation into affecting parameters in laser cutting process is necessary to improve the end product quality. Considerable research studies were carried out to examine the laser cutting process. Yilbas and Sahin (1995) introduced thermal modeling of laser cutting process through accommodating boundary layer effects of the assisting gas. They showed that hydrodynamic boundary layer had signicant effect on the liquid layer thickness around the kerf edges. Yilbas (2001) modeled laser cutting process and stria forma-

tion. He showed that cutting speed and laser power intensity had signicant effect on striation formation around the cut edges. Laser cutting of thick metallic materials was carried out by Allle et al. (1996). They showed that Nd:YAG laser resulted in improved cut quality as compared to CO2 laser cutting. Thick section cutting of stainless steel was examined by Kar et al. (1996). They modeled the cutting process and compared the predictions of the kerf size with the experimental results. They discussed the effects of process parameters, such as laser power, spot size and dimensions, and cutting speed on the resulting kerf size. Numerical and experimental investigations of gas assisted laser cutting of thick metals were carried out by Makashev et al. (1996). They showed that the penetration speed could be improved signicantly through setting laser beam waist position and adjusting gas jet velocity. ONeill and Gabzdyl (2000) investigated laser oxygen assisted cutting of thick metallic substrates. They demonstrated that the immense power could be generated when a controlled ignition process was struck on the surface of a metal plate in the presence of a high-pressure oxygen jet. Laser high power cutting process and formulation of the geometry of the heat affected

E-mail address: bsyilbas@kfupm.edu.sa. 0924-0136/$ see front matter 2007 Elsevier B.V. All rights reserved. doi:10.1016/j.jmatprotec.2007.11.265

286

j o u r n a l o f m a t e r i a l s p r o c e s s i n g t e c h n o l o g y 2 0 1 ( 2 0 0 8 ) 285290

zone were presented by Duan et al. (2001). They considered the effect of various laser process parameters, multiple reections, and inert gas pressure on the geometry of the cut sections. They showed that the shape of the cutting section was strongly dependent on the cutting speed, laser power, and focal position; however, assisting gas pressure had slight effect on the cut geometry. CO2 laser cutting of Inconel alloy and cut quality assessment were carried out by Yilbas (1998). He showed that the laser pulsing frequency had a signicant effect on the end product quality and within certain range of cutting parameters the cut quality improved signicantly. Laser cutting of thick ceramic substrates through controlled fracturing of the irradiated area was investigated by Tsai and Chen (2003). They developed a relationship between the laser parameters and the cut geometry, and stress levels in the cut region were also predicted. The inuence of laser processing parameters on the cut geometry was investigated by Sundar et al. (2005). They identied the key parameter, such as laser pulse frequency, in relation to the resulting cut geometry. The striation formation in laser cutting of metallic materials was examined by Wee and Li (2005). They developed a mathematical model for laser cutting process including the effects of reactive gases. Material removal during laser fusion cutting was modeled by Quintero et al. (2005). They predicted temporal evolution of the material removal mechanisms and the thickness of the molten layer for several laser pulse modes. Striation formation in the laser oxygen assisted cutting was examined by Ermolaev et al. (2006). They predicted the threshold velocity at which a drastic decrease in the size of striations occurred. In the present study, laser cutting of thick sections is considered and modeling of laser cutting process is introduced. Thermal efciency of the cutting process and specic energy required for the cutting are formulated. An experiment is carried out at various levels of the laser parameters and the resulting cut quality is assessed using the international standards for thermal cutting (International Standards). A statistical method employing factorial analysis is introduced to assess the affecting parameters on the cut quality.

Fig. 1 Laser heating situation and coordinate system. Workpiece thickness (w) is 4 mm.

follows: m= d d(wLTh) dL = = wTh dt dt dt (2)

where is the volume of the laser cut hole, is the density of the substrate material, w and L are the width and length of the cut section, and Th is the thickness of the sheet metal. The term dL/dt represents the cutting velocity, which is the laser scanning velocity during the cutting process. In addition, the rate of total loss includes the rate of conduction and convection losses during the cutting, i.e.: Eloss = Econduction + Econvection (3)

where Econduction and Econvection are the rate of conduction and convection losses. The rate of conduction loss from the cutting wall site can be written as: Econduction = d dt
Th 0 0 50w

Cp TL dy

dx

(4)

2.

Mathematical formulation

The specic energy required for a laser cutting of a metallic substrate is formulated using a lump parameter technique. In this case, the rate of energy balance for cutting of a sheet metal is considered in the formulations. The cutting situation and geometric view of the cut section is shown in Fig. 1. The rate of energy required for cutting of a section can be written as: Er = El + m[ u + Lm + Lev ] (1)

where 50w represents the distance far away from the solution domain in the y-direction (radial direction). It is assumed that the natural convection occurs at the top and bottom surfaces of the cut section. The heat transfer coefcients across both surfaces are the same. In this case, the convection losses can be written as: Econvection = As [h1 (Ts T0 ) + h2 (Ts T0 )] = wL (Ts T0 )(h1 + h2 ) (5) where As is the surface area of the cut section, h1 and h2 are the heat transfer coefcient at top and bottom surfaces of the cut section. The specic energy required for the hole cutting is determined from: Espec = 1
tc

where El , m, u, Lm , , Lev are the rate of total losses during the laser cutting, the rate of total mass of material removed during laser cutting, internal energy gain of the substrate material removed, latent heat of melting, fraction of mass evaporated, latent heat of evaporation, respectively. The total material removed during laser cutting, after assuming constant density of the substrate material, is as

Er dt
0

(6)

where tc is the time taken to cut the edges in the sheet metal.

j o u r n a l o f m a t e r i a l s p r o c e s s i n g t e c h n o l o g y 2 0 1 ( 2 0 0 8 ) 285290

287

Table 1 Properties used in the simulations Source of variation

Fraction of evaporation contribution () Latent heat of melting of workpiece Thermal conductivity of workpiece Specic heat capacity of workpiece Thermal diffusivity of workpiece Melting temperature of workpiece

Value
0.1 247112 63 460 1.6134 105 1700

Units
J/kg W/mK J/kg K m2 /s K

Thermal efciency of the laser drilling process is the ratio of the rate of energy required for through cutting over the power used for the cutting process. In this case, thermal efciency becomes: = Ereq P (7)

where P is the laser power on the workpiece surface during the cutting process. The solution of Eqs. (1)(7) are solved using Mathematica software. The data used in the solution are given in Table 1.

which have the main effects while the rst order interactions of the affecting parameters are: laser power cutting speed, laser power pulse frequency, and cutting speed pulse frequency. An F-test is conducted to identify the signicant effects of the parameters and their interactions on the various observations describing the cut quality. The terms most signicant, very signicant and signicant refer to minimum correlation levels of 0.99, 0.990.95, and 0.950.90 of the factors, respectively.

4. 3. Factorial analysis

Experimental

To analyze the effect of laser cutting parameters on the resulting damage size, it is necessary to design an experiment in such a way that the inuence of each parameter could be examined individually in an interaction. This requires a complete factorial design in which selected factors are varied at appropriately selected levels. The percentage of damage size can be estimated by the ratio of damage size around the cut edges to the size of the corresponding cutting edge, i.e.: %Damage size = w w (8)

The laser used in the experiment is a CO2 laser (Mazak, Super Turbo X Mark II) and delivering nominal output power of 4000 W. Oxygen emerging from a conical nozzle and co-axially with the laser beam is used. 127 mm focal lens with defocusing facilities is used to focus the laser beam. The laser output power was ranged 5002000 W during cutting of various workpieces. The laser cutting parameters are given in Table 2. Mild steel sheets with 12 mm thicknesses are cut and the sections cut includes the strait as well as the corners.

5.

Results and discussions

where w is the damage size or enlargement/contraction of the cut edge after the cutting process and w is the cutting edge size (kerf size). It should be noted that damage size and kerf size were measured using an optical microscopy and each measurement was repeated three times to ensure the repeatability of the results. It was estimated that the errors associated with the measurement is within 4%. In statistical analysis, the qualitative factors are nonordered, i.e. each level is of intrinsic interest (Davies, 1978). The mathematical analyses of factorial design are not given here, but refer to Yilbas (1987). It should be noted that in the qualitative analysis, main effects are zero order effects, i.e. independent of each other. However, rst order interactions demonstrate that the effects of the interacted factors on observation are dependent on each other. The affecting factors can be stated as: laser power, cutting speed, and pulse frequency,

Laser cutting of thick workpieces is considered and affecting parameters on the end product quality are investigated. Cutting sections are examined using the optical microscopy and SEM to assess the cut quality. XRD is carried out to determine the oxide compounds formed at the cut surface. Thermal efciency and specic energy requirements for cutting are also analyzed and computed. Fig. 2 shows thermal efciency with laser output power for different cutting velocities. Thermal efciency increases with reducing laser power. This is more pronounced for high cutting velocities. In this case, power required for cutting reduces with increasing velocity, which is more pronounced for high cutting velocities. This is because of the energy required for cutting per unit time reduces with increasing cutting speed. Moreover, as the laser power increases, thermal efciency remains the same, provided that thermal efciency attains high val-

Table 2 Laser cutting parameters Feed rate (m/s)

0.050.5

Power (W)
5002000

Nozzle gap (mm)

1.0

Nozzle diameter (mm)

1.5

O2 pressure (kPa)
450

288

j o u r n a l o f m a t e r i a l s p r o c e s s i n g t e c h n o l o g y 2 0 1 ( 2 0 0 8 ) 285290

Fig. 3 XRD results for laser cut surface. Fig. 2 Thermal efciency of laser cutting with laser power for different scanning speeds.

ues with increasing cutting velocity. In the case of effect of cutting speed on thermal efciency, increasing cutting speed improves thermal efciency signicantly because energy required per unit time increases so that the ratio of energy required per unit time for cutting to laser power intensity increases. It should be noted that although increasing cutting speed while keeping the laser power constant enhances thermal efciency, the cutting process may not be completed due to insufcient laser power for complete cutting. Table 3 gives F-test results obtained from the factorial analysis tests. The percentage of kerf size change is signicantly affected by the laser output power and oxygen pressure; in which case, the main effects of the laser power and oxygen pressure are very signicant. This indicates that increasing laser power increases the depth of the melt zone and increasing oxygen pressure provides increased rate of high temperature oxidation reactions. Consequently, excessive exothermic energy available accelerates thermal erosion from the cut edges. However, the oxidation process is transient in nature and the amount of material thermally agitated from the kerf surface changes. This introduces the melting instabilities at the cut section and magnies the surface irregularity in terms of striations. It is also observed that the rst order interaction of laser power and oxygen pressure is significant. This shows that the coupling effect of laser power and oxygen pressure inuences the percentage of kerf size variation signicantly. Increasing laser power and oxygen pressure enhances the erosion sites in the cut section lowering the cutting quality. It should be noted that when the laser power

Fig. 4 Optical micrograph of top and bottom views of the laser cut section.

and oxygen pressure reduces below the threshold levels, the through cutting ceases and irregular local cutting is resulted. The main effect of the cutting speed inuences the percentage of kerf size, provided that only its effect is signicant. On the other hand, rst order interaction of oxygen pressure and cutting speed is found to be signicant. In this case, lowering cutting speed while increasing oxygen pressure enhances the percentage of kerf size damage. This situation is particularly true for the bottom surface of the workpiece, where the dross is ejected. In the case of cutting speed and laser power intensity, similar arguments are true, which were made for

Table 3 F-test results showing the signicance levels of each parameter and their interactions Source of variations Mean kerf width (w) Kerf width ratio ( w/w) Top surface
Power Cutting speed Oxygen pressure Power cutting speed Power oxygen pressure Cutting speed oxygen pressure 0.95 0.95 0.90 0.90 <0.90 <0.90 0.99 0.95 0.95 0.90 0.90 <0.90

Bottom surface
0.99 0.95 0.95 0.90 0.95 0.90

j o u r n a l o f m a t e r i a l s p r o c e s s i n g t e c h n o l o g y 2 0 1 ( 2 0 0 8 ) 285290

289

Fig. 5 Optical and SEM micrographs of laser cut surfaces.

cutting speed and oxygen pressure. Consequently, increasing laser power while lowering cutting speed increases the thermal erosion from the kerf sites. In addition, increasing laser output power increases the liquid layer thickness due to excessive melting around the cut edges. High temperature combustion taking place in the kerf contributes signicantly to the development of liquid laser around the cut edges. In this case, oxygen-rich iron compounds are formed, such as Fe2 O3 as evident from XRD results as shown in Fig. 3. Moreover, the increase in liquid laser thickness with increasing laser power indicates that early initiation of high temperature oxidation reactions in the cut section enhances the exothermic energy available in the cutting region. This, in turn, enlarges the thermal erosion from the kerf sides and the liquid layer thickness increases signicantly. Fig. 4 shows optical photographs of top and bottom views of laser cut surfaces. The local erosion, excessive melting, and dross attachment are evident from the bottom surface of the workpiece. However, dross attachment and excessive melting are locally scattered around the cut edges. No specic pattern for dross attachment is observed. Fig. 5 shows SEM micrograph of side view of kerf surfaces for different laser power settings and oxygen gas pressures. Deep stria patterns are evident at high laser power intensities and high oxygen pressures. However, increasing cutting speed while decreasing laser power and oxygen pressure reduces the stria depths and stria irregularities appeared at high laser power intensity and oxygen pressures. Moreover, solidication of molten material ow around the cut edges is resulted once the oxygen gas pressure is reduced signicantly. In this case, purging effect of oxygen is reduced considerably towards the kerf exit. Consequently, lowering oxygen pressure further lowers the cut quality. Dross attachment and striation pattern is also evident from Fig. 5. Some small crack sites are observed around the strias at the top surface of the substrate material. This is because of the high cooling rates, which results in high thermal stress eld in this region. However, no specic crack pattern is observed and crack propagates almost in all directions at the surface. This indicates that the cooling is multi-directional and temperature gradient in the region next to the strias is considerably high resulting in multi-cracks with closed spacing.

examined. Thermal efciency and melt layer thickness are formulated and predicted for various cutting parameters. Optical microscopy and SEM are carried out to observe the cut defects and striations around the cut sections. It is found that laser power intensity and oxygen gas pressure had signicant effect on the percentage of kerf width variation. Increasing laser power and oxygen gas pressure results in increased thermal erosion around the cut section; in which case, the depth of stria increases signicantly. However, reducing the cutting speed and oxygen gas pressure results in premature formation of dross on the cut surface; in which case molten metal ow is observed at the cut surface. Moreover, some locally scattered dross attachment is also observed around the cut edges at the bottom surface of the workpiece. Thermal efciency improves at low power intensities and high cutting speeds, provided that once the power intensity is reduced below the threshold through cutting replaces with the deep melting of the workpiece surface.

Acknowledgement
The authors acknowledge the support of King Fahd University of Petroleum and Minerals Dhahran Saudi Arabia.

references

6.

Conclusions

Laser cutting of thick materials is considered and effect of cutting parameters on the percentage of kerf variation is

Allle, J.P., Pilot, G., De Prunele, D., 1996. New pulsed YAG laser performances in cutting thick metallic materials for nuclear applications. Int. Soc. Opt. Eng. 2789, 134144. Davies, O.L., 1978. Design and Analysis of Industrial Experiments, rst ed. Longman Group Ltd., London. Duan, J., Man, H.A., Yue, T.M., 2001. Modelling the laser fusion cutting process III. Effects of various process parameters on cut kerf quality. J. Phys. D Appl. Phys. 34, 21432150. Ermolaev, G.V., Kovalev, O.B., Orishich, A.M., Fomin, V.M., 2006. Mathematical modelling of striation formation in oxygen laser cutting of mild steel. J. Phy. D: Appl. Phys. 39, 42364244. International Standards. DIN 2310, West Germany. Kar, A., Scott, J.E., Latham, W.P., 1996. Theoretical and experimental studies of thick-section cutting with a chemical oxygen-iodine laser (COIL). J. Laser Appl. 8, 125133. Makashev, N.K., Buzykin, O.G., Asmolov, E.S., 1996. Computational and experimental investigation of gas-assisted laser cutting of thick metal. Proc. SPIEInt. Soc Opt. Eng. 2713, 248252. ONeill, W., Gabzdyl, J.T., 2000. New developments in laser-assisted oxygen cutting. Opt. Lasers Eng. 34, 355367.

290

j o u r n a l o f m a t e r i a l s p r o c e s s i n g t e c h n o l o g y 2 0 1 ( 2 0 0 8 ) 285290

Quintero, F., Varas, F., Pou, J., Lusquinos, F., Boutinguiza, M., Soto, R., Perez-Amor, M., 2005. Theoretical analysis of material removal mechanisms in pulsed laser fusion cutting of ceramics. J. Phys. D: Appl. Phys. 38, 655666. Sundar, J.K.S., Thawari, G., Sundararajan, G., Joshi, S.V., 2005. Inuence of process parameters during pulsed Nd:YAG laser cutting of nickel-base superalloys. J. Mater. Process. Technol. 170, 229239. Tsai, C., Chen, H., 2003. Laser cutting of thick ceramic substrates by controlled fracture technique. J. Mater. Process. Technol. 136, 166173. Wee, L.M., Li, L., 2005. An analytical model for striation formation in laser cutting. Appl. Surf. Sci. 247, 277284.

Yilbas, B.S., 1987. Study of affecting parameters in laser drilling of sheet metals. Trans. ASME J. Eng. Mater. Technol. 109, 282287. Yilbas, Z., 1998. Classication of striation patterns using a neural network in CO2 laser cutting. Laser Eng. 7, 2537. Yilbas, B.S., 2001. Effect of process parameters on the kerf width during the laser cutting process. Proc. Inst. Mech. Eng. Part C: J. Eng. Manuf. 215, 13571365. Yilbas, B.S., Sahin, A.Z., 1995. Oxygen assisted laser cutting mechanicsa laminar boundary approach including the combustion process. Opt. Laser Technol. 27, 175184.