WO2001028734A1

WO2001028734A1 - Improvements in laser machining

Info

Publication number: WO2001028734A1
Application number: PCT/GB2000/003887
Authority: WO
Inventors: Andrew Kearsley; Martyn Knowles
Original assignee: Oxford Lasers Limited
Priority date: 1999-10-16
Filing date: 2000-10-11
Publication date: 2001-04-26
Also published as: GB9924419D0; GB2355222B; GB2355222A

Abstract

A method is disclosed of machining a hole through a workpiece (3) using a laser the workpiece (3) having a thickness (T) at least ten times the minimum diameter of the laser beam (1) at its focus (F), the method comprising the steps of: selecting a laser which produces a laser beam (1) with a Gaussian or other soft edged profile; passing the laser beam (1) through a first optical component (4) which modifies the beam (1) so as to output a beam (1) with a substantially hard-edged profile; positioning the laser beam (1) relative to the workpiece (3) such that the focus (F) of the beam (1) lies part-way through the thickness (T) of the workpiece (3); and causing the beam (1) to move around a closed pathway on the workpiece (3) so as to form a hole through the workpiece (3) the shape of which is determined by the shape of the closed pathway.

Description

IMPROVEMENTS IN LASER MACHINING

FIELD OF THE INVENTION

This invention relates to a method of machining a hole through a workpiece using a laser, to a laser for use in the method and to an optical component for use in the method.

PRIOR ART

In the machining of materials with lasers, the effect of the laser on the material depends strongly on the power density at the material surface. A number of regimes can be identified:

a) Very low power density. When the power density is very low the material is heated relatively slowly and heat spreads into the surrounding material so that the irradiated area and the material surrounding this is heated but does not melt. This heat can affect the material causing changes to its crystal structure and this can be detrimental to the component being manufactured.

b) Low power density. When the power density is low, the material is heated slowly to above the melting point but significant vaporisation does not take place. Heat also spreads into the surrounding material. After the laser pulse terminates, the molten material re-solidifies, possibly with the formation of oxidised material.

c) Moderate power density. At moderate power densities, there is significant melting accompanied by vaporisation of some of the material. There is some ejection of material in the form of particulate lumps or droplets of a wide size range caused by the pressure of the vapour. After the end of the laser pulse, material which had been melted but not ejected, re-solidifies. In some cases, the re-solidified material can form irregularly shaped bolsters or tongues around the pit formed at the laser target spot. This material may contain a high proportion of oxide and slag.

d) High power density. At high power densities, the energy delivered to the volume defined by the area of the laser target spot (and the surrounding heat diffusion zone) and the absorption depth of the laser beam (again possibly increased by heat diffusion and/or mechanical shock) is enough to melt and vaporise the material in this volume many times over. In this case, the material is explosively ejected from the pit in the form of small particulates or droplets and vapour. Little or no material remains to re-solidify around the lip of the pit thus created although, in the case of metals, some oxide and slag may be deposited as a weakly adhering layer on surrounding surfaces as the vapour is scattered in the air above the target.

It is this last regime that is the most effective and desirable for the operation of laser machining, since the material is ejected cleanly with the least effect on the surrounding bulk material.

A laser beam from a laser of good beam quality (e.g. with m squared less than 5), may be brought to focus at a target spot whose diameter is of the order of the wavelength of the laser multiplied by the f number of the lens used to converge the beam multiplied by the m value of beam. For lenses typically used in laser machining of materials more than a few microns in thickness, and with lasers of wavelength in the visible or near IR (infra-red) region, this implies that the laser spot will be no smaller than about five microns in diameter and typically in the range 10 to 20 microns in diameter. The effective f number is the focal length of the lens divided by the beam diameter at the lens.

M squared is a times-diffraction-limit-factor that relates beam divergence of the laser to the diffraction limited divergence, e.g. m squared = 2 means that the laser beam divergence is twice the diffraction limit.

In order to form holes (or other features) with dimensions of around ten microns or larger, one possible strategy would be to de-focus the laser target spot to the full diameter of the desired final hole and to subject the material to a train of laser pulses covering this area - the so called impact (or percussion) laser drilling approach. A disadvantage of this strategy is that the power density at the laser target spot is lower than the value which could be achieved if the beam were more tightly focussed, and may correspond to the situations described in b) and c) above rather than the more desirable case d). Alternatively, the pulse energy of the laser could be increased to maintain intensity. However, the technique requires the beam to have the exact shape and intensity profile required to form the hole. This is very difficult to achieve and gets progressively more difficult as the power is increased. However, despite its shortcomings, the percussion approach is normally favoured in laser drilling - for example when drilling cooling holes in turbine blades.

A preferred strategy when laser drilling high precision holes in thin materials (e.g. less than 50 micron thick) is to adopt the trepanning approach. In this case, the laser beam is brought to focus at its target spot on the material at or close to the minimum achievable diameter. This ensures that machining will be carried out at power densities corresponding to the desirable regime of case d) above. To form holes of diameter larger than that of the laser spot, the position of the target spot on the surface of the workpiece is caused to trace out a path on the workpiece in the form of a circle (or whatever cross-sectional shape is desired). The translation of the laser target spot across the surface of the workpiece can be brought about by either translating the workpiece relative to a fixed laser beam or, more conveniently, by allowing the workpiece to remain stationary and deflecting the beam optically so as to cause the target spot to describe the desired path. This technique enables a much higher precision profile to be formed than can be achieved by creating the exact shape and size with the beam.

When the trepanning approach is used, the material is either cut through completely from laser beam entry side to the beam exit side before the trepanning motion brings the target spot back to its initial position on the workpiece after one circuit, or it may take several circuits to achieve breakthrough.

In a preferred implementation of the trepanning strategy, a clear hole is initially created in the material of a diameter (or characteristic dimension) smaller than desired in the finished item. The trepanning process is then continued so that the outermost region of the laser beam slightly overlaps the edge of the hole thus formed. The edge material is thus rapidly vaporised and removed, and the process is continued so the diameter of the trepanned hole thus formed is gradually increased to the desired finished size. This has the added benefit of removing the majority of the ejected material that may have been deposited on the surface around the hole.

In conventional methods of trepanning, the laser beam is arranged so the intensity across a cross-section through the beam, particularly at its focus, follows a substantially Gaussian profile in order to optimise the shape and form of the laser spot on the workpiece. The present invention aims to provide an improved method and apparatus for machining a hole in a workpiece by trepanning through thick workpieces. In this context, 'thick' means a workpiece having a thickness at least ten times the diameter of the laser beam at is thinnest point, i.e. at its focus. A typical laser beam may have a minimum diameter of around 5 microns in which case a ^'thick^' workpiece is one having a thickness of at least 50 microns. In this situation, the workpiece has a thickness which is greater than the depth of focus of the laser beam.

DISCLOSURE OF THE INVENTION

According to a first aspect of the invention, there is provided a method of machining a hole through a workpiece using a laser, the workpiece having a thickness at least ten times the minimum diameter of the laser beam at its focus, the method comprising the steps of: selecting a laser which produces a laser beam with a Gaussian or other soft edged profile; passing the laser beam through a first optical component which modifies the beam so as to output a beam with a substantially hard-edged profile; positioning the laser beam relative to the workpiece such that the focus of the beam falls part-way through the thickness of the workpiece; and causing the beam to move around a closed pathway on the workpiece so as to form a hole through the workpiece the shape of which is determined by the shape of the closed pathway.

According to another aspect of the invention, there is provided a laser for use in such a method, the laser being arranged to produce a laser beam with a Gaussian or other soft-edged profile and having an optical component through which the laser beam is passed and which is arranged to produce an output laser beam having a substantially hard-edged profile. According to a further aspect of the invention, there is provided an optical component for use in such a method, the optical component comprising a pair of elements each element having a substantially wedge-shaped cross-section arranged such that a laser beam passed therethrough is deflected around a closed pathway as the pair is rotated about an axis, the relative angular orientation of the elements about the axis determining the diameter of the pathway.

The invention also relates to a workpiece having a hole machined therethrough by such a method.

Preferred and optional features of the invention will be apparent from the following description and from the subsidiary claims of the specification.

A hard edged beam is a beam in which the power density falls off rapidly at the edge of the beam from a high value to a negligible value. In contrast, in a soft edged beam, the power density falls off gradually at the edges of the beam, such as in a beam in which the power density across the beam follows a Gaussian profile.

BRIEF DESCRIPTION OF THE DRAWINGS

Figure 1 is a schematic diagram illustrating the use of a focussed laser beam for machining a hole in a thick workpiece and shows different profiles the laser beam may have, including a profile used in one aspect of the invention;

Figure 2 is a schematic diagram of optical apparatus used to deflect a laser beam during a trepanning process according to another aspect of the invention; and Figure 3 illustrates a spiral path along which the laser beam may be moved by the apparatus shown in Figure 2.

BEST MODE OF CARRYING OUT THE INVENTION

In thick materials as defined above (e.g. having a thickness greater than 50 microns) a problem is encountered with the conventional trepanning process. In lasers with a high beam quality, the power fall off with absolute radial distance is fairly rapid at the focus of the beam as the beam is very small at the focus. However, when machining materials with a thickness which is greater than the length of the beam 'waisf , i.e. its focal depth, the radial profile of the laser beam becomes important. In a typical machining situation, it may be necessary to machine a hole through a workpiece 1 mm thick. In this case, the depth of focus of the beam will be only a few tens of microns so most of the material will be machined by light that is out of focus.

In this example, if the machining is performed with the focus of the laser set to fall in approximately the middle of the material, then the upper and lower surfaces of the workpiece see a beam which is 0.5 mm out of focus. In a typical set up, with an f=10 lens, this means that the beam has a diameter of approximately 50 microns at the surface of the workpiece.

A laser beam with an approximately Gaussian profile maintains a Gaussian profile as it nears the focus and a beam 50 microns in diameter with a Gaussian profile has a 1/e radius of about 25 microns. However, significant power is still present outside this, e.g. at a radius of 50 microns. It is therefore inevitable that a large area exists that is irradiated with a power density as described by regimes a), b) and c) described above. As a result, a poor quality hole is produced. According to one embodiment of the invention, this problem can be overcome by using a laser beam that has a profile of an approximately "top hat" shape. When such a beam is focussed, it maintains a sharp edge throughout its depth of focus until it nears the focal plane where diffraction effects dominate. A 50 micron diameter beam having a "top hat" profile has a radius of 25 microns and beyond this it falls from full power to negligible power over a distance of just a few microns. As a result the amount of material in power density regimes a), b), and c) is minimised so a good quality hole can be formed. Such a beam is referred to as having a 'hard' edge, that is to say, in the region of the target spot, the power density falls off as rapidly as possible from a high value within the core of the beam to negligible values at the beam edge.

Figure 1 illustrates the above and shows a laser beam 1 , a focussing lens 2 and a workpiece 3 in which a hole is to be machined. The thickness T of the workpiece is at least 50 microns.

Figure 1 illustrates a first situation in which a beam 1A having a substantially Gaussian profile, as shown by curve G1 , is incident upon a surface 3A of the workpiece 3. The beam 1A is focussed so the focal point, or the beam "waist", is located approximately half-way through the thickness of the workpiece. The profile at the focal plane F is also Gaussian as shown by curve G2.

In the method described herein, the beam 1 is given a hard surface by inserting an aperture plate 4 in the beam 1 so as to block the soft edges thereof. The beam 1 B passing through the aperture in the plate 4 thus has a top hat profile with hard edges when it is incident upon the surface 3A of the workpiece, as shown by curve H1. This top hat profile is maintained until the beam approaches the focal plane F and diffraction effects dominate as shown by curve H2. The aperture plate 4 may be positioned outside the laser but, in some cases an aperture plate, or other optical component which performs a similar function, can be positioned within the laser chamber itself.

It is particularly important in the final process of reaming the hole that the laser beam should have a hard edge. A gradual decrease of power density outwards from beam centre is not desirable as the yet un-machined material of the wall of the hole should either be subjected to laser beam power densities of the magnitude of case d) above, or to no power at all. To subject them to the low power densities typical of the outer radial regions of a 'soft-edged' beam (such as a Gaussian profile) runs the risk of producing the undesirable heat damage associated with the power regimes a), b) or c) without producing the clean cut desired.

In prior art methods, the use of a hard-edged beam would generally be avoided as diffraction effects tend to enlarge the minimum spot size that can be achieved compared to a beam with a Gaussian profile. However, it has now been realised that, when machining through a 'thick' workpiece, this disadvantage is counteracted by the advantage of being able to control the beam profile at the surface of the workpiece.

The present invention thus makes use of the properties of a hard-edged beam to enable a high quality hole to be machined by minimising the opportunity for power density regimes a), b) or c) to arise.

A further consideration arises when machining with pulsed lasers. The pulse always has a finite rise time and fall time. While the maximum power of the pulse may be sufficient to ensure that the majority of the material is irradiated with light as in the power regime d) described above, the power will fall into the unwanted regimes at the start and end of the pulse. In particular, the pulses from many solid state lasers often have a long trailing 'tail' . While the main part of the pulse may have a 30 ns full width, half maximum duration (i.e. its width at half maximum power) it can often have a tail which is over 100 ns long. This long tail can again causes unwanted heating and damage to the surrounding material.

Thus, for machining holes in material where the thickness is larger than the depth of focus of the laser beam, it is an advantage to use a laser beam which has a hard edge rather than a beam with a Gaussian profile. In addition, it is desirable to use a laser which produces a pulse which rises from zero power and falls to zero power as quickly as possible.

The laser pulse should preferably be such that the power of the pulse falls to less than 5% of the peak power value within 100 ns of the peak, and most preferably to less than 1 %.

One application that has received attention recently is the laser drilling of holes in diesel injector nozzles. A typical nozzle requires a number of holes with a diameter in the range 40 to 200 microns through material 0.5 to 1.5 mm thickness. This falls into the "thick" regime described above.

A number of attempts have been made to drill nozzles with solid state lasers (e.g. Nd:YAG laser with fundamental wavelength of 1.06 micron, or frequency doubled to 532nm, or similar solid state lasers). Even when the best beam quality is used (typically from a diode pumped laser) poor quality holes result. However, by using the techniques described above, i.e. by using a beam profile with a hard edge instead of a Gaussian or near Gaussian profile and by ensuring the laser pulse duration is not so long that the "tail" of the laser pulse causes heat damage, the quality of the holes can be significantly improved. Thus, in a preferred embodiment of the invention, a trepanning process is carried out with a hard edged beam and a laser which produces a pulse which has a very short fall time. For example, a high beam quality, frequency doubled, diode pumped Nd:YAG laser can be used. This normally produces a near Gaussian beam profile but can be given a hard edge by the insertion of an aperture plate in the beam to remove the lower power parts. While this reduces the power in the beam and may increase the spot size at the focus to some extent (due to diffraction effects), it results in a beam which maintains a "hard" edge away from the focus. If the use of an aperture is combined with the selection of a laser with a short pulse (e.g. falling to less than 5% of peak power within 100ns of the peak), and used in a high speed trepanning process the process is capable of forming high quality holes. The laser beam is preferably moved around the trepanning pathway at a frequency of at least 5 Hz, and most preferably at least 10 Hz.

To aid the achievement of the required short pulse duration from the laser, frequency multiplication is preferably used. As this is a non-linear process, dependent on the square of the power, the output pulses are typically significantly shorter for visible or ultra-violet (UV) wavelengths than they are for infra-red (IR) wavelengths. This also has the advantage that in most materials visible or UV wavelengths are more strongly absorbed than IR wavelengths. If visible or UV wavelengths are used, it is thus desirable to ensure that negligible IR light is present in the main visible or UV beam as IR light will be brought to focus in different places and the IR beam will have a longer tail than the visible or UV beam.

Furthermore, when a frequency multiplied laser is used, the laser beam is usually polarised so the polarisation state of the beam incident on the workpiece is dependent upon the orientation of the part of the pathway being irradiated. This is a disadvantage because the absorption of the light depends on whether the polarisation is parallel to or perpendicular to the wall. As the beam is rotated, the polarisation changes from parallel to perpendicular every 90° of rotation and this can lead to deviations from the desired hole profile. This problem can be overcome by using a quarter-wavelength (λ/4) plate to create circular polarised light or by rotating a half-wavelength (λ/2) plate to rotate the polarisation incident on all parts of the hole wall.

In order to trepan very precise circular holes, it is possible either to move the workpiece in an accurate circle or to move the beam. Because of the masses involved it is normally not practical to move the workpiece and usually the beam is moved around a circular orbit. Various methods can be used including rotation of an off axis lens, reflection off spinning mirrors, reflection off a pair of galvanometer mounted mirrors and the rotation of prisms. Of these the most stable system uses rotating prisms.

A beam of light passed through a first prism with a small wedge angle is deviated by a small amount. To first approximation, if the light hits the prism at an angle close to the minimum deviation angle (i.e. approximately normal to the surface of a small angle prism) then the deviation is independent of the point on the prism the light hits and on the angle the face of the prism makes to the light. Hence, if the prism moves or tilts relative to the beam, the direction of the light does not change. However, if the prism is rotated, the beam will describe a very accurately defined cone even if the rotation mount is not of perfect quality. If the cone of light is focused by a lens, the focus will describe a very accurate circle. The diameter of the circle described is given by 2Φf where Φ is the deviation angle and f the focal length of the lens.

A refinement of this is to use a pair of prisms that can rotate independently. If they are constrained to rotate with a fixed relative angle, they effectively form a single prism with a wedge angle that depends on the relative angle between the wedges. Hence holes of arbitrary size up to 4Φf can be formed using such a pair of prisms.

A further refinement is to use a pair of prisms with very small wedge angles (typically less than 1 degree) and to rotate them at high speed (e.g. above 10 Hz). When combined with the hard-edged laser described above this enables very accurate small holes to be machined. In addition, by independently controlling the rotation of the two prisms, the relative angle therebetween can be changed during the machining process, so enabling the reaming process described above to be carried out. For even higher accuracy two dissimilar prisms can be used. One has a wedge angle that will form a hole of approximately the correct size without a second prism being used. The second prism has an even smaller wedge angle and acts a vernier to give very fine control over the size of the hole being machined.

In another refinement, more than two prisms can be used to allow additional control or to allow the machining of non-circular holes. For example, four such prisms can be used to produce elliptical holes.

Figure 2 shows an embodiment of such an arrangement for deflecting the laser beam around a circle. The optical component shown comprises a pair of elements in the form of wedge-shaped prisms 5 and 6 mounted for rotation about an axis A. A first motor 7 is used to rotate prism 5 about axis A and a second motor 8 is used to rotate prism 6 about axis A. The motors 7,8 are connected to the respective prisms 5,6 by precision drive belts (not shown). The relative angular orientation of the prisms 5,6 about the axis A and their speed of rotation about axis A are precisely controlled by operation of the motors and rotary encoders 7,8, e.g. in the form of disks with marks or slots on them mounted on the motor shafts. Rotation of the disk is monitored by sensing movement of the marks or slots optically and so can be monitored and controlled by electronic control means such as a computer 9. Thus by using appropriate software, rotation of the prisms 5 and 6 can be very accurately controlled and can easily be modified to suit the requirements.

This apparatus deflects the beam about a circular path, but by arranging for the relative angular orientation of the two prisms 5,6 about the axis A to be gradually changed, the beam can by moved along a spiral path, as shown in Figure 3, before reaching the circular path of the required diameter. A laser beam can thus be used to first pierce the workpiece at the centre of the circle and then follow the spiral path as it is gradually moved out to the circular path of the required diameter.

In an alternative arrangement, a single motor may be used to rotate both (or all) prisms and gears used to govern the relative rotational speeds of each prism.

Other solid state lasers may be used in place of Nd:YAG laser. A carbon dioxide laser may also be modified to provide a laser beam with a hard edge for use in a method as described above.

The method described above is particularly suited to the laser machining of holes in workpieces such as: ink jet printer nozzles, diesel injector nozzles, injector nozzles for petrol internal combustion engine, via holes in printed circuit boards (i.e. holes for connecting circuits on one side of the board with the other side of the board) and cooling holes in jet turbine engine components e.g. a turbine blade or engine combustor, as in all these applications the workpiece falls within the 'thick' regime described above and very accurate holes are required.

Claims

1. A method of machining a hole through a workpiece using a laser the workpiece having a thickness at least ten times the minimum diameter of the laser beam at its focus, the method comprising the steps of: selecting a laser which produces a laser beam with a Gaussian or other soft edged profile; passing the laser beam through a first optical component which modifies the beam so as to output a beam with a substantially hard-edged profile; positioning the laser beam relative to the workpiece such that the focus of the beam lies part-way through the thickness of the workpiece; and causing the beam to move around a closed pathway on the workpiece so as to form a hole through the workpiece the shape of which is determined by the shape of the closed pathway.

2. A method as claimed in claim 1 in which the workpiece has a thickness of at least 50 microns.

3. A method as claimed in claim 1 or 2 in which the first optical component comprises an aperture plate arranged to block transmission of soft edges of the laser beam.

4. A method as claimed in claim 1 , 2 or 3 in which the laser is a solid state laser.

5. A method as claimed in claim 4 in which the solid state laser is a Nd:YAG laser.

6. A method as claimed in claim 1 , 2 or 3 in which the laser is a carbon dioxide laser.

7. A method as claimed in any preceding claim in which laser pulses are used, the power of the pulses falling to less than 5%, and preferably less than 1%, of the peak power value within 100 ns of the peak.

8. A method as claimed in any preceding claim in which the laser beam is moved around said closed pathway at a frequency of at least 5 Hz and preferably at least 10 Hz.

9. A method as claimed in any preceding claim in which the laser is frequency is multiplied.

10. A method as claimed in claim 9 in which the laser produces visible or UV light.

11. A method as claimed in any preceding claim in which the laser beam is passed through a second optical component which modifies the polarisation state of the beam so that the polarisation state of the beam incident on the workpiece is not dependent upon the orientation of the pathway.

12. A method as claimed in claim 11 in which the second optical component is a quarter-wavelength plate or a rotating half-wavelength plate.

13. A method as claimed in any preceding claim in which the laser beam is passed through a third optical component which is arranged to deflect the beam so that it moves around the closed pathway on the workpiece.

14. A method as claimed in claim 13 in which the third optical component is arranged to move the beam along a substantially spiral path before moving if around said closed pathway.

15. A method as claimed in claim 13 or 14 in which the third optical component comprises a pair of elements each having a substantially wedge-shaped cross-section, the relative angular orientation of the two elements about an axis determining the diameter of the closed pathway as the pair is rotated about said axis.

16. A method as claimed in claim 14 and 15 in which the beam is moved along the substantially spiral pathway by adjusting the relative angular orientation of the two elements about said axis as the pair is rotated about said axis.

17. A method as claimed in any preceding claim used to machine holes in one of the following workpieces: an ink jet printer nozzle; an injector nozzle for a diesel internal combustion engine and an injector nozzle for a petrol internal combustion engine; via holes in a printed circuit board and cooling holes in a jet turbine engine component.

18. A method of machining a hole through a workpiece substantially as herein before described and/or with reference to the accompanying drawings.

19. A workpiece having a hole machined therethrough by a method as claimed in any preceding claim.

20. A laser for use in a method as claimed in any of claims 1 to 18, the laser being arranged to produce a laser beam with a Gaussian or other soft- edged profile and having an optical component through which the laser beam is passed and which is arranged to produce an output laser beam having a substantially hard-edged profile.

21. A laser as claimed in claim 20 arranged to produce a pulsed laser beam in which the power of the pulses fall to less than 5%, and preferably less than 1%, of the peak power value within 100 ns of the peak.

22. A laser as claimed in claim 20 or 21 in combination with an optical component arranged to deflect the beam so that it moves around a closed pathway.

23. A laser as claimed in claim 22 in which the optical component comprise a pair of elements, each having a substantially wedge-shaped cross-section such that the relative angular orientation of the elements about an axis determines the diameter of the pathway as the pair is rotated about said axis.

24. An optical component for use in a method as claimed in any of claims 1 to 18, the optical component comprising a pair of elements each element having a substantially wedge-shaped cross-section arranged such that a laser beam passed therethrough is deflected around a closed pathway as the pair is rotated about an axis, the relative angular orientation of the elements about the axis determining the diameter of the pathway.

25. An optical component as claimed in claim 24 comprising a first motor for rotating a first element of the pair about said axis and a second motor for rotating the second element of the pair about said axis.

26. An optical component as claimed in claim 25 comprising control means for controlling the relative angular orientation of the two elements and their speed of rotation about said axis by controlling the operation of the first and second motors.

27. An optical component in which the control means comprises a computer.

28. A laser as claimed in claim 20 substantially as hereinbefore described with reference to and/or as shown in the accompanying drawings.

29. An optical component as claimed in claim 24 substantially as hereinbefore described with reference to and/or as shown in the accompanying drawings.