WO2009007708A2

WO2009007708A2 - Laser cutting

Info

Publication number: WO2009007708A2
Application number: PCT/GB2008/002338
Authority: WO
Inventors: Mohamed Sobih; Lin Li; Philippus Crouse
Original assignee: The University Of Manchester
Priority date: 2007-07-09
Filing date: 2008-07-08
Publication date: 2009-01-15
Also published as: WO2009007708A3

Abstract

A method of laser cutting, the method comprising generating a laser beam and directing the laser beam and assist gas at a workpiece, relative movement between the laser beam and the workpiece taking place such that the laser beam scans over a surface of the workpiece, wherein parameters including the power of the laser beam, the spot size of the laser beam in the workpiece, the pressure of the assist gas, and the cutting speed are selected such that the workpiece is cut without causing striations to be formed on surfaces of the cut.

Description

LASER CUTTING

The present invention relates to laser cutting.

Laser cutting has been used in industry to cut materials such as metal for over 35 years, and is a widely accepted technique for sheet metal profiling. When laser cutting is used, periodic lines known as striations are formed on surfaces which are cut by the laser. These striations add roughness to the cut surfaces, thereby reducing the quality of the surfaces. In addition, the striations cause variation of the width (kerf) of the cut provided by the laser.

It is an object of the present invention to provide improved laser cutting.

According to a first aspect of the invention there is provided a method of laser cutting, the method comprising generating a laser beam and directing the laser beam and assist gas at a workpiece, relative movement between the laser beam and the workpiece taking place such that the laser beam scans over a surface of the workpiece, wherein parameters including the power of the laser beam, the spot size of the laser beam in the workpiece, the pressure of the assist gas, and the cutting speed are selected such that the workpiece is cut without causing striations to be formed on surfaces of the cut.

Preferably, the power of the laser beam and the spot size of the laser beam in the workpiece are selected to provide a beam power density which is sufficient that pushing of molten material out of the cut by the vaporisation front is the dominant material removal mechanism in the workpiece.

Preferably, the power of the laser beam, the spot size of the laser beam in the workpiece, and the cutting speed are selected such that the time taken for a vaporisation front generated by the laser beam to advance from the top to the bottom of the workpiece is equal to or less than the beam-material interaction time of the laser beam and the workpiece. Preferably, the power of the laser beam, the spot size of the laser beam in the workpiece, and the cutting speed are selected such that the equation

is satisfied, where d is the average diameter of the laser beam in the workpiece material, v_max is the maximum cutting speed of the laser cutting, h is the thickness of the workpiece, and v_v is the vaporisation speed of the workpiece during laser cutting.

Preferably, the assist gas is oxygen, and the cutting speed is sufficiently fast that an oxidation front caused by the cutting does not escape from the laser beam.

Preferably, the cutting speed is selected such that it satisfies the following equation:

v nu ■n τt ~ ^τc where v_min is the minimum cutting speed of the laser cutting, d is the laser beam diameter, τ_t is the oxidation reaction termination time and τ_c is the time at which oxidation front is caught by the laser beam.

Preferably, the cutting speed is selected such that it satisfies the following equation: r_rvi_in -rfv_πώl -2Z) = 0 where v_min is the minimum cutting speed of the laser cutting, d is the laser beam diameter, τ_t is the oxidation reaction termination time and D is the diffusion coefficient of the migrating species through the oxide layer.

Preferably, the beam diameter at a lower surface of the workpiece is less than the kerf width at an upper surface of the workpiece.

Preferably, the parameters are selected according to the equation:

where v_max is the maximum cutting speed of the laser cutting, /yis the reflectivity of the material to the laser beam is the laser beam reflectivity of the workpiece material, F is the power density of the laser beam, d is the average diameter of the laser beam in the workpiece material, p is the material density of the workpiece material, h is the thickness of the workpiece material, L_v is the latent heat of vaporisation of the workpiece material, C_p is the thermal capacity of the workpiece material, T_b is the boiling temperature of the workpiece material, and ΔH₀₂ is the exothermic reaction energy of the workpiece material.

Preferably, the assist gas is provided at a pressure which is sufficiently high that a desired proportion of the workpiece material is oxidised, but not so high that cooling which causes striation formation in the cut occurs.

Preferably, the pressure of the assist gas is within the range 0.5 to 2.3 bar.

Preferably, a standoff distance between the workpiece and a delivery nozzle which delivers gas to the cut is within the range 0.7 to 2.4 mm.

Preferably, the focus of the laser beam is within the range of 1.7 to 5.7 mm above an upper surface of the workpiece.

Preferably, the power density of the laser beam at the workpiece is greater than 3 x 10⁶ W/cm².

Preferably, the cutting speed is in the range 57 to 124 mm/s.

Preferably, the half angle of laser beam divergence after the focal point is less than 2 . Preferably, the half angle of laser beam divergence after the focal point is less than

0.675

According to a second aspect of the invention there is provided a laser cutting apparatus, the apparatus comprising a laser arranged to generate a laser beam, collimating optics arranged to collimate the laser beam, a laser cutting head arranged to focus the laser beam and to deliver assist gas, and a table arranged to support a workpiece onto which the laser beam is directed, relative motion being possible between the table and the laser beam such that the laser beam scans over a surface of the workpiece, wherein parameters including the power of the laser beam, the spot size of the laser beam in the workpiece, and the cutting speed are set such that the workpiece will be cut by the laser without causing striations to be formed on surfaces of the cut.

The laser cutting apparatus of the second aspect of the invention may be configured or arranged to operate according to any of the preferred features of the first aspect of the invention.

According to a third aspect of the invention, there is provided a material which has been cut according to the first aspect of the invention.

According to a fourth aspect of the invention there is provided a material with a striation-free cut achieved using laser cutting.

A specific embodiment of the invention will now be described by way of example only with reference to the accompanying drawings, in which:

Figure 1 is a schematic illustration of a laser cutting a workpiece according to an embodiment of the invention; Figures 2 to 4 are graphs which illustrate the progression of oxidation reactions during laser cutting;

Figures 5 to 7 are schematic illustrations of oxidation reactions during laser cutting;

Figure 8 is a graph which illustrates the progression of oxidation reactions during laser cutting;

Figures 9 and 10 are pictures which compare striation-free cutting achieved using cutting parameters according to an embodiment of the invention, with cutting achieved using cutting parameters which fall outside of the embodiment of the invention; and

Figures 1 1 to 14 are graphs which illustrate the effect of changing parameters of the laser cutting.

In an embodiment of the invention a 1 kW maximum power output continuous wave IPG YLR-1000-SM ytterbium single-mode fibre laser is used to cut a workpiece. The laser has a 14 μm delivery fibre core diameter, and is configured to generate a laser beam having a TEMoo beam intensity distribution with a wavelength of 1.07 μm and an M² factor of 1.1. The laser is commercially available from IPG Photonics of Massachusetts, USA.

Figure 1 shows schematically the laser 2 generating a laser beam 4 which is incident upon a workpiece 18. The laser beam generated by the laser is collimated to a diameter of 3mm using a pair of lenses 6. Following collimation, the laser beam is focussed using a focussing lens 8 having a focal length of 5 inches (127 mm). The diameter of the laser beam spot in the focal plane 10 of the focussing lens 8 is 32 μm, and beam divergence (half angle) before the focal point is 0.675°.

The focussing lens 8 is provided within a laser cutting head 12, and is moveable in the z-direction within the laser cutting head. The laser cutting head 12 used is a HP 1.5" (Z)ZFL cutting head with Lasermatic® Z distance measurement system, available from Precitec Inc of Michigan, USA. The term HPl. 5" indicates that the diameter of the focussing lens in the cutting head is 1.5 inches (38 mm). The cutting head includes capacitive distance sensors that constantly record the standoff distance between the cutting nozzle and the workpiece. An electronic unit analyses the sensor signal, and uses this to control a linear drive. The laser cutting head has an exit hole 14 with a diameter of 1 mm (this is exchangeable, allowing an exit hole of for example 0.5, 1.5, or 2 mm to be used instead). The exit hole is provided in a coaxial conical nozzle, which is used to deliver gas to the cutting location. The gas, which is provided via an input 16 in the laser cutting head, has a pressure which may be set to between 0.5 and 16 bars using a regulator. The gas, which will hereafter be referred to as assist gas, may be O₂ or any other suitable gas. Although the cutting head has a focal length of 5 inches (127 mm), the focussing lens within the cutting head may be exchanged for a focussing lens with a different focal length, for example 7.5 inches (191 mm) or 10 inches (254 mm). When a focussing lens with a focal length of 191 mm is used, then the diameter of the laser beam spot in the focal plane 10 of the focussing lens 8 is 48 μm, and beam divergence (half angle) before the focal point is 0.3°.

The laser cutting head 12 delivers the laser beam 4 onto the workpiece 18. In this case the workpiece is a 2 mm thick annealed EN43 mild steel sheet. The workpiece is held on a computer numerically controlled (CNC) x-y table (i.e. a table with a surface which may be moved in the x and y directions using a computer). The laser beam 4 is not moveable in the x and y directions. The laser cutting head 12 is however moveable in the z-direction. The laser cutting head is typically positioned between 0.5 and 2 mm from the workpiece. The distance between the laser cutting head 12 and the workpiece 18 is referred to as the nozzle standoff distance.

Striation-free cutting of the mild steel sheet 18 may be achieved using the apparatus described in relation to figure 1. However, striation-free cutting is only achieved if appropriate operating parameters are selected.

Although the generation of striations during laser cutting is a well known phenomenon, there is no consensus amongst those skilled in the art regarding the processes which give rise to striations. Similarly, there is no consensus regarding how to suppress the formation of striatums. Indeed, the commonly held view is that striations will always occur (striation-free cutting has never before been achieved).

A formula has been devised which may be used to determine, for a given material to be cut, operating parameters which will provide striation-free cutting of that material.

In conventional laser cutting, the material removal mechanism is based on "melt and blow", i.e. a laser beam melts the material and a coaxial gas jet removes the material. In contrast to this in the present invention, the principal mechanism via which melted material is removed from the cut during striation-free cutting is vaporisation of the material (the pressure of the assist gas is sufficiently low that it has a negligible effect on material removal). The laser beam gives rise to a vaporisation front, which advances downwards through the workpiece during cutting. In other words, vaporisation begins at the upper surface of the workpiece, and passes down through the workpiece (i.e. in the z-direction) to the lower surface of the workpiece. As the vaporisation front passes downwards through the material it pushes molten material out of the cut. It is this pushing of molten material out of the cut by the vaporisation front which is understood to be the main cause of material removal from the cut. The vaporisation front has a vaporisation front speed, which is the speed at which the vaporisation front advances downwards through the workpiece. The time taken for the vaporisation front to pass from an upper surface of the workpiece to a lower surface of the workpiece depends upon the speed of the vaporisation front and the thickness of the workpiece.

The time taken for the laser beam to advance in the direction of cut (e.g. x-direction) over a distance which is equivalent to the laser beam diameter is known as the beam- material interaction time. If the time taken for the vaporisation front to advance from the top to the bottom of the workpiece is less than or equal to the beam-material interaction time, then the vaporisation front has sufficient time to pass fully downwards through the material. Pressure from the vaporisation front pushes melted material rapidly downwards. The vaporisation front travels downwards so quickly that the laser drives melted material out of the cut before the melted material can solidify on the sidewall of the cut (solidified material would give rise to striations). Thus, striation free cutting may be achieved if sufficient time is available to for the vaporisation front to pass fully downwards through the material during cutting. This defines a maximum cutting speed for striation free cutting. If a cutting speed higher than this were to be used, then the vaporisation front would not pass fully downwards through the material, and striations would be seen in the cut.

Since vaporisation takes place across the entire depth of the cut, a high temperature (thus low viscosity) melt film is maintained across the depth surrounding the laser beam. Although assist gas may be used, striation free cutting may be achieved when the pressure of the assist gas is low enough (1-2 bar) that its contribution to material removal is negligible. The vaporisation front advancement in the axial direction (i.e. along the beam, down the cut kerf) may therefore be considered as the driving force for material removal (the vaporisation front pushes molten material out of the cut). Therefore, if the time taken for the vaporisation front to travel through the depth of the workpiece is less than or equal the time of laser/material interaction within the beam spot, and assuming melting to be a continuous process (i.e. for a continuous wave laser beam), melt oscillation can be prevented (and hence striation formation may be prevented).

The vaporisation front speed can be estimated from Carslow and Jaeger's equation: m (l - r, )F _{(1 )}

" P(L_v + C_pT_b) where v_v is the vaporisation front speed, r^is the reflectivity of the material to the laser beam, F is the power density of the laser beam, p is the material density of the workpiece material, L_v is the latent heat of vaporisation of the workpiece material, C_p is the thermal capacity of the workpiece material, and T_b is the boiling temperature of the workpiece material. In order to determine the maximum cutting speed at which striation-free cutting may be achieved, the following should be satisfied:

J- = JL (2) where d is the average laser beam spot diameter through the depth of the workpiece, h is the workpiece thickness, and v_max is the maximum cutting speed at which striation- free cutting occurs (cutting speed refers to the distance travelled per second across the surface of the workpiece). Combining equations (1) and (2) and considering a contribution of energy from O₂ reaction gives the maximum cutting speed for striation-free cutting:

where a_o is the percentage of contribution of O₂ reaction to the total energy density (a_o = 0 for an inert gas). Assuming that the O₂ reaction occurs through a cylinder of material interacting with the laser beam over the depth of the material, and that in general half of the material is oxidized [Ivarson et al., 1991, The role of oxidation in laser cutting stainless and mild steel, Journal of Laser Applications, 3(3): 41-45], then:

F₀, - ¹^"' (4) where v is the cutting speed, F₀^ is the equivalent exothermic reaction power density, ΔH_θ2 is the exothermic reaction energy, and d is the average laser beam diameter through the depth of the cut. Therefore

FQ₁ _ 2≠vΔH_Oi ^

«o₂ =

(l - r,)F nd(\ - r_f )F

Equation 3 can thus be expressed as:

or solving for v_max: n{\ - r_f)Fd V^max ph[π{L_{v +} C_pT_b)-2AH_θ2 ] ^U

Taking the example of a 2 mm thick annealed EN43 mild steel sheet, the following material values are used in equation 7: p = 7880 kg/m³, L_v=6.1xl0⁶ J/kg, T_b = 3023K, C_p=460 J/(kgK), h = 2xlO^"3 m, r/ = 0.4 (material reflectivity measured using a spectrometer from Ocean Optics of Florida, USA), Δ//_θ2 = 4.79 x 10⁶ J/kg (for an

FeO dominated reaction product). The following laser beam parameters, which pertain to the IPG YLR-IOOO-SM ytterbium single-mode fibre laser, are used: average spot size d = (116 + 167)/2 μm = 1.4IxIO^"4 m (i.e. the average of the beam diameter through the steel sheet; 116 μm is the beam diameter at the top of the workpiece and 167 μm is the beam diameter at the bottom of the workpiece), average power density

F= 6.38x10¹⁰ W/m² (for 1000 W laser power).

Putting all of the above values into equation 7 gives an estimated maximum speed for striation-free cutting of v_max = 11 mm/s for the 2 mm thick mild steel sheet. The assumptions included in the calculation mean that an accuracy of +/- 10% should be applied to this value. For example, if the reaction product is not FeO dominated then the value of ΔH_θ2 may be different, which would in turn affect the value of v_max.

Equation 7 may be used to determine a combination of average spot size, average power density and maximum cutting speed which may be used to achieve striation free cutting of any suitable material.

The minimum cutting speed at which striation free cutting may be achieved arises from the speed at which an oxidation front propagates across the material. The oxidation front will occur when oxygen is used as the assist gas. When cutting mild steel oxygen is typically used, because it substantially increases the achievable cutting speed. This is because the oxygen causes a reaction which releases energy, and therefore speeds up the cutting.

The cutting speed must be greater than or equal to a minimum value such that the oxidation front is not able to escape from (i.e. propagate away from) the laser beam. The minimum cutting speed at which striation free cutting may be achieved is the speed at which cyclic side burning generated by the oxidation reaction is eliminated. The cyclic nature of the oxidation reaction during laser cutting can be explained by the time dependence of the oxidation reaction speed. The dependence of the thickness of the oxide layer with respect to time is parabolic, i.e.

z² = IDt . (8)

Here D is the diffusion coefficient of the migrating species through the oxide layer (m²/s) and t is the time (s). Rearranging equation 8 and taking the derivative with respect to time, an expression is obtained for the velocity of the oxidation front, i.e.

Here v_ox is the oxidation speed (m/s), i.e. the speed at which the oxidation front moves across the material. The diffusion coefficient of iron in the oxide layer is fiver times greater than that of oxygen, and so the diffusion of iron through the oxide layer is the dominant transport process. The diffusion coefficient of iron in oxide (=3.5x10^~6 m²/s) is used, instead of the oxygen diffusion (=6.6x10^"7 m²/s).

At the beginning of the reaction, the oxidation speed is very high and the oxidation front escapes from the laser beam. However, equation 9 indicates that the oxidation speed is inversely proportional to the square root of time, as shown in figure 2 for the diffusion coefficient of iron in oxide. This results in a decrease of the heat generated from the oxidation reaction as time increases. Consequently the oxidation reaction will terminate after a certain time, T_h as the heat generated is no longer adequate to sustain the reaction, and the movement of oxidation front stops.

When the laser beam catches up with the stationary oxidation front, a new reaction is initiated. The oxidation front escapes from the laser beam and then the oxidation reaction dies once more. In this manner the process repeats cyclically.

In an alternative situation, since the speed of the oxidation front speed decreases with time, the laser beam may catch up with the oxidation front before the oxidation reaction has died. This would occur at a time τ_c, the time at which both the laser beam and the oxidation reaction have travelled the same distance. This time can be calculated using the following equation, which is derived by inverting equation 8, i.e.

ID τc = -^r (10) where D is the diffusion coefficient of the migrating species through the oxide layer and v is the cutting speed.

To calculate the reaction termination time an energy balance equation is solved. The process energy balance can be written as

P_l + P_o, = P_m + K + P_loss - (H)

The left hand side represents the heat input, and this is equal to the heat output on the right hand side. Here P_/ is the laser power, P_0x is the power released from the oxidation reaction, P_n, is the power consumed to melt a portion of the metal, P_v is the power consumed to vaporize the rest of metal and Pι_oss is the power losses due to conduction and convection. Because the following is concerned only with the oxidation reaction an assumption is made that the laser only initiates the reaction, and the contribution of the laser is not included in the calculation. The quantity of material that is vaporized is small, and so the power consumed in vaporization is also neglected. Only the heat loss due to conduction P_cond is considered. Equation 11 thus simplifies to

or

P_m - P_m = P_cond - (13)

The power released from the oxidation reaction is

P_ox ^ 4→z)v_oxpAH_θ2 (14)

where h is the workpiece thickness, d is the laser beam diameter, z is the position of the oxidation front calculated from Equationδ, v_ox is the speed at which the oxidation front moves across the material, p is the workpiece material density and AH₀, is heat released from the oxidation reaction (=4.82x10⁶ J/kg of Fe). The power consumed to melt the workpiece is

[ 2 OX p(AH 300K- + AH₅₁ ) (15)

where Δ"_{300A: >τ} is the enthalpy required to raise the workpiece material temperature from room temperature to the melting point (=65 kJ/mole of Fe) and AH_5/ is the enthalpy of melting the workpiece material (=19.4 kJ/mole of Fe). The conduction heat loss in the solid material from a cylindrical heat source φ

where q_mcs is the heat flux at the boundary of a moving cylinder source and q_mcs(^r><P) = ^k(^Tm ^{~ T}o)^Pe exp(-Pe^'r cos<p)

Here k is the thermal conductivity, r is the laser beam radius, φ is the azimutal angle, T_m is the melting point, To is the room temperature, /, ,Ki are the modified Bessel functions of first, second and i^th order and Pe, Pe are forms of the Peclet number defined as

Pe = -^V≡^ (18) 2a

Pe = ^■ (19)

2a where v is the scanning speed of the cylindrical heat source (oxidation front speed), and a is the thermal diffusivity.

The reaction termination time (τ,) at which P_0x - P_m = P_cond can be calculated using conventional mathematical software to solve Equation 13 in the form /(O = P_0x (t) - P_m (0 - P_cmd (0 = 0 , to get the root which is τ, . This may be done using the definitions of P_ox(t), P_m(t) and P_COnd(t) provided in Equations 14, 15 and 16. An example of mathematical software which may be used is Mathcad, available from Parametric Technology Corporation of Massachusetts, USA.

solving equations 14, 15 and 16 numerically, the reaction termination time (T₁) at which P_0x - P_m - Pcond can be calculated, shown in figure 3, for a beam diameter d=140 μm and using the thermal properties of mild steel. This results in τ_t = 4.323 ms. Equation 10 can be used to calculate the time (r_c) at which laser beam catches the oxidation front at different cutting speeds. Neither τ_t nor T_c depends on the workpiece thickness according to this approach.

In figure 4 the termination time (τ_r) at which the energy balance holds, and the time at which the laser beam catches up with the decelerating oxidation front (τ_c) in the direction of cut before the reaction is terminated, are plotted. The two plots coincide at about 40 mm/s. The cutting process below this velocity is schematically illustrated in figure 5 (in this case the velocity is 30 mm/s). Up to this point of coincidence, the calculated time required to catch the oxidation front is higher than the time at which oxidation reaction terminates, so the oxidation front stops at time (r₍) as shown in figure 5c. The laser beam catches up with the stationary oxidation front as shown in figure 5d, and starts a new cycle as shown in figures 5e to 5g.

At cutting speeds higher than this value, two subsequent scenarios may occur, as depicted in figure 6 and figure 7 respectively. In figure 6 the cutting speed is 50 mm/s. After the initial ignition, the oxidation front escapes from the laser beam, as shown in figure 6b. After that at time (r_c) the laser beam catches up with the oxidation reaction only at the leading edge, as shown in figure 6c, while the oxidation front is still in motion, and the two fronts effectively move together. This is understood to be a pseudo steady state. During this period the laser power term should be introduced into the energy balance equation as it no longer negligible. However, lateral oxidation is still independent of the laser power and remains powered by the oxidation enthalpy. The lateral oxidation reaction stops at time equal to (T_t), as shown in figure 6e. When the laser beam passes through re-ignition points it ignites the sides of the oxidation front, and a new oxidation reaction cycle starts. This is shown in figure 6f and g. Two effective oxidation fronts escape from the laser, as shown in figure 6f. These oxidation fronts will give rise to striations. Thus striations occur due to lateral cyclic oxidation, steady state conditions being maintained only at the leading edge.

Comparing the time difference between the time at which both the laser beam and the oxidation reaction have travelled the same distance (r_c) and the time when the side way oxidation reaction terminates (τ_t) on one hand, and the interaction time

r,. = - (20) v on the other hand, it is observed that at a given cutting speed, T_rT_c is higher than the interaction time. This situation is illustrated in figure 7, in which the cutting speed is 65 mm/s. Physically this means that a true steady state reaction is achieved here and the oxidation front in this case cannot escape from the laser beam, either in the forward or in the lateral directions. The cutting process is now controlled by the laser beam, resulting in steady state cutting, thereby allowing striation-free cutting, as shown in figure 7. In other words, since the oxidation front is unable to escape from the laser, striations which would be caused by the propagating oxidation front no longer occur. From figure 8, the point of intersection between T_rτ_c and interaction time (Ti) graphs is at a cutting speed of nearly 57 mm/s. This value is understood to be the minimum cutting speed at which striation free cutting may be achieved (based on the values set out above). The value is believed to have an accuracy of +/- 10%, and is the lower limit of the operating window obtained experimentally (experimental results are shown further below). The value, which will be referred to as v_min, may be represented as:

V^_n = -^- , (21)

where d is the laser beam diameter, τ, is the time at which the oxidation reaction will terminate (can be calculated using equation 13), and τ_c is the time at which both the laser beam and the oxidation reaction have travelled the same distance (can be calculated using equation 10). Substituting for τ_c from equation 10 and rearranging, this can be expressed as:

where D is the diffusion coefficient of the migrating species through the oxide layer.

Equation 7 and equation 21 (or equation 22) together define the maximum and minimum cutting speeds between which striation free cutting may be achieved. As has been explained, the striations formed at cutting speeds below the minimum value arise from a mechanism which is entirely different from the mechanism which causes striations to form at cutting speeds above the maximum value. Experimental results show that striations formed at low speeds are different in form from striations formed at high speeds, thus support this understanding.

In order to achieve striation free cutting of 2 mm thick mild steel, the cutting speed used should be between 57 mm/s and 77 mm/s (+/- 10%), as explained above. For other materials, or for mild steel of a different thickness, equations 7 and 21 can be used to determine the minimum and maximum cutting speeds respectively.

Two additional conditions should be satisfied for striation-free cutting of the 2 mm thick mild steel sheet: the gas pressure should be less than 2.3 bars, and the beam diameter within the cut kerf across the cutting depth should be small enough to sustain vaporisation. These conditions ensure the dominant material removal mechanism is pushing of molten material out of the cut by the vaporisation front (an assumption mentioned above). The conditions also ensure that the melted material has a high temperature and thus a low viscosity. A low viscosity is desirable because it allows pressure from the vaporisation front to push melted material downwards (as explained above), thereby removing the material from the cut and suppressing sideways propagation of melting (which may cause striation).

In conventional laser cutting, a laser beam with substantial beam divergence is used. This means that at the diameter of the laser beam at the bottom of the cut is greater than the width of the kerf at the top of the cut. hi the embodiment of the invention the beam divergence is low, and so the beam diameter throughout the cut remains less than the width of the kerf at the top of the cut (i.e. the beam diameter at the bottom of the cut is less than the width of the kerf at the top of the cut).

The pressure of the assist gas is relevant because it has an influence on the proportion of material in the cut zone which is oxidised, hi the above calculation, 50% oxidation of the material in the cut zone is assumed. If, for example, O₂ assist gas is used then a higher gas pressure will increase the proportion of material in the cut zone which is oxidised. This may help to cause vaporisation of the material, and may thus be of assistance in obtaining striation free cutting. However, if the pressure of the assist gas is too high, it will cause cooling of the material, and this may be detrimental to the striation-free cutting process. Therefore, a balance is needed. The gas pressure should be sufficiently high that a desired proportion of the material is oxidised, but not so high that cooling occurs which is detrimental to the cut. In the experimental set-up described above, the gas pressure should be less than 2.3 bars, as confirmed from figure 13. However, other gas pressures may be appropriate for other experimental setups - e.g. different laser beam power density, different stand-off distance, etc.

The power density of the laser beam should be sufficient to cause vaporisation of the material to be cut. In the embodiment of the invention the wavelength of the laser beam is 1.07 μm. Other wavelengths may be used. However, the wavelength will affect the beam focal spot size, and hence the achievable power density. This should be taken into account when considering using a different wavelength. In addition, absorption of the laser beam by the workpiece material will have a wavelength dependency (in general longer wavelengths are less well absorbed). This criteria should also be considered when selecting a laser beam wavelength.

In the embodiment of the invention the M² factor is 1.1 (M² is a measure of the quality of the laser beam). M² affects the focusability and depth of focus of the laser beam. A small M implies that the laser beam can be easily focused to small spot size, thereby providing a high power density. It is not necessary that the M² factor is 1.1 in order to achieve striation free cutting. However, the M² factor should allow focussing which gives a power density of the laser beam sufficient to cause vaporisation of the material, hi addition, the M factor should allow the beam diameter at the bottom of the cut to be less than the width of the kerf at the top of the cut.

It is not necessarily the case that striation-free cutting always arises with the smallest laser beam spot size. Too small a beam spot size will result in a very short beam- material interaction time. This will mean that a higher vaporisation front speed is needed. In addition it will provide a smaller kerf width. The smaller kerf width may restrict assist gas and ejected material from flowing into and out of the kerf. Where this occurs, striation free cutting may in some instances not be achieved. The embodiment of the invention uses a single mode fibre laser which is capable of delivering a beam spot size of 15-70 μm. This is smaller than is achievable using conventional CO₂ lasers (100-250μm) or Nd: Y AG lasers (80-600 μm). Therefore, a defocused beam is used in this example to achieve striation-free laser cutting (in the example above the beam spot size is 116 μm at the upper surface of the workpiece).

Cutting experiments have been performed, to demonstrate combinations of parameters which will provide striation free cutting.

A range of cutting experiments were performed in which the laser power was varied between 600 and 1000 W and the cutting speed varied between 30 and 125 mm/s (cutting speed refers to the distance travelled per second across the surface of the workpiece). The position in the z-direction of the focus of the laser beam 4 was varied between -1 mm and 15 mm (i.e. from 1 mm below the upper surface of the workpiece to 15 mm above the upper surface of the workpiece). The z-position of the focus was adjusted by moving the focussing lens 8 in the z-direction. The nozzle standoff distance was varied between 0.5 and 3 mm.

The laser cut samples were analysed using optical microscopy. The laser cut samples were also analysed using a Veeco-Wyko NTI lOO optical surface profiling system (available from Veeco Instruments Inc of New York, USA), which measures both surface profile and roughness.

Figure 9 shows the result of using the apparatus described in relation to figure 1 (including the IPG YLR-1000-SM ytterbium single-mode fibre laser) to cut a 2 mm thick annealed EN43 mild steel sheet, as seen through an optical microscope. Figure 9a shows a cut which was achieved when the laser beam had a power of IkW, and the cutting speed was 30 mm/s. Striations can clearly be seen in the surface of the cut. Figure 9b shows a cut which was achieved when the laser beam again had a power of IkW, and a cutting speed of 70 mm/s was used. Striations are not present on the surface of the cut. The cut is therefore considered to be striation-free.

Figure 10 shows the result of using the laser cutting apparatus to cut a 2 mm thick annealed EN43 mild steel sheet, as viewed using the optical surface profiling system. The scale on the right hand side of the pictures shown in figure 10 indicates height/depth variation in μm with respect to a calculated average height. Figure 1 Oa shows a cut which was achieved when the laser beam had a power of IkW at a cutting speed of 30 mm/s. Striations of the cut surface are present. Figure 10b shows a cut which was achieved when the laser beam again had a power of IkW, but with a cutting speed of 80 mm/s. Striations are not present on the cut surface.

Although some surface roughness can be seen in the cut achieved using a cutting speed of 80 mm/s, this is substantially smaller than the roughness seen when using a cutting speed of 30 mm/s, and does not have the periodic variation which is characteristic of striations. The cut is therefore considered to be striation-free.

Experiments have been performed to identify the effect of cutting speed and laser power on the formation of striations. Figure 11 a shows the effect of cutting speed on striation formation at different laser powers for 1 mm thick mild steel using 5" focal length lens, stand off distance 1 mm, oxygen pressure 1 bar and focal plane set 4 mm above the workpiece surface. Figure 1 Ib shows the effect of cutting speed on striation formation at different laser powers for 2 mm thick mild steel, with the same focal length lens, standoff distance, oxygen pressure and focal plane position. The results show operating windows within which striation free cutting is achieved.

Figure 11 shows that striation decreases as the cutting speed increases until striation free cutting is achieved. When cutting 2 mm thick mild steel a minimum laser power of around 680 Watts is needed. No such minimum laser power is seen when cutting 1 mm thick mild steel (for the range of powers over which the experiments were performed).

In general, increasing the power of the laser increases the range of speeds over which striation free cutting may be achieved. In particular, a generally linear dependency of the cutting maximum speed versus laser power is seen (as would be expected based upon equation 7

At high cutting speeds, when cutting 2 mm thick mild steel striations occur once more. At still higher speeds the steel is not cut. When cutting 1 mm thick mild steel, there is a transition directly from striation free cutting to no cut.

The laser beam used to generate the results shown in figure 11 had a diameter of 1 16 μm at the upper surface of the workpiece. This means that for example 800 W of laser power corresponds with a power density of 7.6 x 10⁶ W/cm² at the workpiece upper surface, and 1000 W of laser power corresponds with a power density of 9.5 * 10⁶ W/cm² at the workpiece upper surface. Additional results are shown in figure 12. The results shown in figure 12 were achieved with cutting head 12 delivering O₂ assist gas at 1 bar pressure, and with the cutting head 12 at a standoff distance of 2 mm from a 22 thick mild steel sheet (standoff distance refers to the distance between the exit hole 14 of the cutting head and the upper surface of the workpiece).

The term 'striation free cutting' is intended to mean cutting in which the roughness of the cut is less than 2.2 Ra (μm) and the periodic lines which characterise striations (see for example figures 9a and 10a) are no longer seen. The term 'Ra' refers to the arithmetical mean deviation of the assessed profile. Based on this understanding of striation free cutting, it can be seen from figure 12 that striation free cutting of 2mm thick mild steel was achieved over a range of cutting speeds from approximately 60mm/s to 80mm/s.

Figures 13a and 13b show the effect of varying the standoff distance between the exit hole 14 and the workpiece, and of varying the oxygen pressure. In figure 13a a 1 mm thick piece of mild steel is cut using 5" focal length lens, with a laser power of 1000 W, a cutting speed of 70 mm/s and the focal plane set at a distance of 4 mm above the workpiece surface. In figure 13b a 2 mm thick piece of mild steel is cut (other parameters remain the same). Both figures show operating windows within which striation free cutting is achieved. For the 1 mm thick piece of mild steel the operating window may be said to roughly be centred on a stand off distance of 1.6 mm and an oxygen pressure of 1.6 bar. For the 2 mm thick piece of mild steel the operating window may be said to roughly be centred on a stand off distance of 1.6 mm and an oxygen pressure of 1.5 bar.

Figure 14 shows the effect of varying the position of the focus of the laser beam (i.e. varying the position of the beam waist), and varying the pressure of the oxygen. Figure 14a is for a 1 mm thick steel workpiece, whereas figure 14b is for a 1 mm thick steel workpiece. The results were obtained using the 5" focal length lens, with a laser power of 1000 W, a cutting speed of 70 mm/s and a stand off distance of 1 mm. Both figures show operating windows within which striation free cutting is achieved. It can be seen that the operating windows are relatively small, compared with the ranges of values over which measurements were obtained. For the 1 mm thick piece of mild steel the operating window may be said to roughly be centred on a focal plane position of 4 mm above the surface of the workpiece, and an oxygen pressure of 1.1 bar. For the 2 mm thick piece of mild steel the operating window may be said to roughly be centred on a focal plane position of 4.8 mm above the surface of the workpiece and an oxygen pressure of 1.2 bar.

Figures 9 to 14 show that by selecting appropriate operating parameters, striation-free cutting of 1-2 mm EN43 mild steel sheets can be achieved. The operating parameters needed to achieve striation free cutting are very different from the operating parameters used in conventional laser cutting, e.g. using a CO₂ or an NdrYAG laser. The operating parameters which are used in conventional laser cutting, and some operating parameters which have been used to obtain striation free laser cutting of mild steel are summarised in table 1 :

Table 1

It can be seen from table 1 that striation-free laser cutting conditions are outside of the normal laser cutting regimes. The reduced gas pressure and increased standoff distance mean that the gas velocity and pressure at the workpiece is lower than in conventional laser cutting. It has previously been noted that reduced gas pressure will reduce striation depth (see for example Chen, K., Yao, Y.-L., 1999, Striation Formation and Melt Removal in Laser Cutting Process, Journal of Manufacturing Systems, 1 : 43-53. ). However, it has not previously been suggested that striation free cutting is achievable.

The focus of the laser beam is 4-5 mm from the surface of the workpiece, and this means that the kerf width is wider than it would have been if the focus was aligned with the upper surface of the workpiece (the kerf width is 350 μm, whereas it would have been 100 μm if the focus were placed on the workpiece).

Due to the small beam diameter (3 mm) at the focussing lens 8, the beam divergence from the beam focus is small. Therefore, the laser beam diameter within the cutting depth can be kept small (the term 'small' is intended to mean smaller than would be the case in conventional laser cutting). In conventional laser cutting, the laser beam has a diameter of 15-30 mm before it is focussed, and it therefore has a high divergence (7-20°). This means that although the upper surface of the workpiece is in the focal plane and receives a focussed laser beam, the beam diverges quickly such that the lower part of the workpiece receives a much broader laser beam. In contrast to this, the laser beam used by the embodiment of the invention has a lower divergence. The divergence may be less than 2° after the focal point, and may be for example 0.675° or less. The low divergence of the laser beam used by the embodiment of the invention allows a high power density to be maintained across the cutting depth (i.e. from the upper surface to the lower surface of the workpiece). This is important because it assists in creating a high temperature zone in which pushing of molten material out of the cut by the vaporisation front is the dominant material removal mechanism. In general, the power density used by the embodiment of the invention is 8-10 times greater than the power density used by conventional laser cutting. In the embodiment of the invention pushing of molten material out of the cut by the vaporisation front is the dominant material removal mechanism (i.e. removes the majority of the material).

Taking the example used above, for a 2 mm thick workpiece placed 4 mm below the focal plane, the beam diameter may be 1 16 μm at the top surface and 167 μm at the bottom surface of the workpiece respectively. For a laser power of 700-1000 W, this gives a power density of 6.6 x 10⁶ - 9.5 x 10⁶ W/cm² on the top surface of the workpiece and 3.2 x 10⁶ - 4.6 x 10⁶ W/cm² at the bottom of the workpiece (if internal reflections within the cut kerf are not considered), hi general, this can be expressed as saying that the power density in the workpiece should be more than 3 x 10⁶ W/cm² for 2 mm thick mild steel (the value being expressed to the nearest whole number).

Typical beam material interaction times along the beam centre line at cutting speeds of 60-75 mm/s are 1.6-1.9 ms on the top surface and 2.2-2.8 ms at the bottom of the workpiece. Under these conditions, vaporisation takes place across the entire depth of the cut. This assists in maintaining a high temperature (thus low viscosity) melt film across the workpiece area which surrounds the laser beam during cutting. As mentioned above, a contribution of energy from O₂ reaction may arise (i.e. from oxidation). This means that increasing the assist gas pressure may allow the cutting speed to be increased whilst maintaining striation free cutting. However, as mentioned above, increasing the gas pressure too much may cause cooling of the workpiece, and thereby cause striation to occur.

The energy contribution from O₂ reaction may be seen when cutting Iron or Titanium based materials. When cutting stainless steel or ceramics oxygen should not be used as the assist gas. Other gases such as nitrogen or argon may be used.

Although the embodiment of the invention has been described in relation to an IPG YLR-IOOO-SM ytterbium single-mode fibre laser, it will be appreciated that other lasers may be used, provided that they are capable of providing a laser beam with sufficient power density (i.e. with a sufficiently narrow focus, and a sufficiently small divergence).

In the described embodiment of the invention, the laser beam is fixed in x and y, and a CNC table is used to move the workpiece relative to the radiation beam. However, the laser beam itself may move relative to the workpiece. Indeed, both the laser beam and the workpiece may be moveable.

Although the embodiment of the invention relates to cutting mild steel, the invention may be used to cut any other suitable material (for example Titanium based material, stainless steel, ceramics, etc). The explanation above of the minimum cutting speed, which arises from the speed of the oxidation front, does not apply when there is no oxidation front present. This is the case when cutting materials without using oxygen as the assist gas.

Claims

1. A method of laser cutting, the method comprising generating a laser beam and directing the laser beam and assist gas at a workpiece, relative movement between the laser beam and the workpiece taking place such that the laser beam scans over a surface of the workpiece, wherein parameters including the power of the laser beam, the spot size of the laser beam in the workpiece, the pressure of the assist gas, and the cutting speed are selected such that the workpiece is cut without causing striations to be formed on surfaces of the cut.

2. The method of claim 1, wherein the power of the laser beam and the spot size of the laser beam in the workpiece are selected to provide a beam power density which is sufficient that pushing of molten material out of the cut by the vaporisation front is the dominant material removal mechanism in the workpiece.

3. The method of claim 2, wherein the power of the laser beam, the spot size of the laser beam in the workpiece, and the cutting speed are selected such that the time taken for a vaporisation front generated by the laser beam to advance from the top to the bottom of the workpiece is equal to or less than the beam-material interaction time of the laser beam and the workpiece.

4. The method of any of claims 1 to 3, wherein the power of the laser beam, the spot size of the laser beam in the workpiece, and the cutting speed are selected such that the equation

5. The method of any preceding claim, wherein the assist gas is oxygen, and the cutting speed is sufficiently fast that an oxidation front caused by the cutting does not escape from the laser beam.

6. The method of claim 5, wherein the cutting speed is selected such that it satisfies the following equation:

v nu-n = ^■ τ_t - τ_c where v_min is the minimum cutting speed of the laser cutting, d is the laser beam diameter, τ_t is the oxidation reaction termination time and τ_c is the time at which oxidation front is caught by the laser beam.

7. The method of claim 5, wherein the cutting speed is selected such that it satisfies the following equation:

where v_min is the minimum cutting speed of the laser cutting, d is the laser beam diameter, τ, is the oxidation reaction termination time and D is the diffusion coefficient of the migrating species through the oxide layer.

8. The method of any preceding claim, wherein the beam diameter at a lower surface of the workpiece is less than the kerf width at an upper surface of the workpiece.

9. The method of any preceding claim, wherein the parameters are selected according to the equation:

10. The method of any preceding claim, wherein the assist gas is provided at a pressure which is sufficiently high that a desired proportion of the workpiece material is oxidised, but not so high that cooling which causes striation formation in the cut occurs.

11. The method of claim 9, wherein the pressure of the assist gas is within the range 0.5 to 2.3 bar.

12. The method of any preceding claim, wherein a standoff distance between the workpiece and a delivery nozzle which delivers gas to the cut is within the range 0.7 to 2.4 mm.

13. The method of any preceding claim, wherein the focus of the laser beam is within the range of 1.7 to 5.7 mm above an upper surface of the workpiece.

14. The method of any preceding claim, wherein the power density of the laser beam at the workpiece is greater than 3 x 10⁶ W/cm².

15. The method of any preceding claim, wherein the cutting speed is in the range 57 to 124 mm/s.

16. The method of any preceding claim, wherein the half angle of laser beam divergence after the focal point is less than 2°.

17. The method of claim 16, wherein the half angle of laser beam divergence after the focal point is less than 0.675°.

18. A laser cutting apparatus, the apparatus comprising a laser arranged to generate a laser beam, collimating optics arranged to collimate the laser beam, a laser cutting head arranged to focus the laser beam and to deliver assist gas, and a table arranged to support a workpiece onto which the laser beam is directed, relative motion being possible between the table and the laser beam such that the laser beam scans over a surface of the workpiece, wherein parameters including the power of the laser beam, the spot size of the laser beam in the workpiece, and the cutting speed are set such that the workpiece will be cut by the laser without causing striations to be formed on surfaces of the cut.