US3448240A

US3448240A - Method of processing workpieces by means of energy-carrying rays

Info

Publication number: US3448240A
Application number: US637251A
Authority: US
Inventors: Karl Heinz Steigerwald
Original assignee: Individual
Current assignee: Messer Griesheim GmbH
Priority date: 1966-05-14
Filing date: 1967-05-09
Publication date: 1969-06-03
Anticipated expiration: 1986-06-03
Also published as: SE334800B; FR1503075A; GB1114985A

Description

June 3, 1969 K. H. STEIGERWALD 3 ,44

METHOD OF PROCESSING WORKPIECES BY MEANS OF ENERGY-CARRYING RAYS Filed May 9, 1967 Sheet Of 6 lN VENTOR. KARL HEINZ STEIGERWALD KZMLI'MMWL ATTORNEYS METHOD OF PROCESSING WORKPIECES BY MEANS OF ENERGY C/XRRYING RAYS Sheet Filed May 9, 1967 v F/G.3b FIG. 30 FIG. 30' FIG. 3e

INVENTOR. KARL HEINZ STEIGERWALD ATTORNEYS June 3, 1969 K. H. STEIGERWALD 3,448,240

METHOD OF PROCESSING WORKPIECES DY MEANS OI" ENERGY CARRYING I'm-KY5 Filed May 9. 1967 Sheet 3 of 6 l FIG. 7

FIGS

FIG. 10 58 lNVENTOR. KARL HEINZ STEIGERWALD ATTORNEYS June 3, 1969 K.-H. STEIG'ERWALD 3,448,240

METHOD OF PROCESSING WORKPIECES BY MEANS OF ENERGYCARRYING HAYS Filed May 9. 1967 Sheet L of e m 62 FIG. 12

I FIG. 14

INVENTOR. KARL HEINZ STE IGERWALD ATTORNEYS June 3, 1969 K H.ISTEIGERWAL.D 3,448,240-

METHOD OF PROCESSING WORKPIECES BY MEANS OF ENERGY-CARRYING RAYS Filed May 9, 1967 Sheet 5 01 6 FIG. 77a FIG. 77b

FIG. 1a

a2 84 85 ee V 79 a3 as 91 FIGZO INVE NTOR. KARL HEINZ STEIGERWALD ATTORNEYS United States Patent 3,448,240 METHOD OF PROCESSING WORKPIECES BY MEANS OF ENERGY-CARRYING RAYS Karl Heinz Steigerwald, Haderunstr. 1a,

Munich, Germany Filed May 9, 1967, Ser. No. 637,251 Claims priority, application Germany, May 14, 1966, St 25,390 lint. Cl. B23k /00 US. Cl. 219-121 32 Claims ABSTRACT OF THE DISCLOSURE The method of welding includes the step of first forming an entrance or inlet opening to substantially the depth of the desired weld seam, introducing a high energy ray into the opening and moving the ray from the entrance opening along the weld seam to weld the material. Various welding details to overcome change of hole size as the weld progresses is described.

This invention relates to an improved method of welding using a high energy ray or beam.

Machining of workpieces, such as cutting, welding, drilling and milling, by means of high energy beams (e.g., laser or charge carrier beams such as electron beams) is known to the art.

When welding and cutting metal workpieces of a relatively great thickness, it is difficult to obtain an adequate penetration depth by the rays. To overcome this difiiculity when welding, it has been proposed to use a sharply focused ray and to make its intensity so high that the ray iotmiaioreljhe workpiece, at the point of impact by melting and evaporating the material through which the ray energy penetrates deeply into the workpiece. This technique for welding has become known as deep ray welding and is disclosed, for example, in U.S. Patent No. 2,987,610. In quite a number of materials, it has become possible to produce very narrow weld seams, the technical properties of which are 'better than those of weld seams produced by way of other known welding methods. This is attributed to the fact that in deep ray welding, the energy requirement per unit of surface area to be welded is lower than in conventional welding methods. The reactions of material and workpiece occuring in each welding methods, more particularly distortion, shrinking, Welling out and spraying out of the molten material from the weld seam, mechanical stresses, thermally caused changes of the workpiece structure and composition and so on, are consequently considerably fewer when using deep ray welding than in other known welding methods.

In such deep ray welding, however, adverse features are encountered which, in many cases, render a technically satisfactory and reproducible welding difiicult or even impossible. One serious disadvantage of known deep ray welding consists in that the material at the point of impact has to be heated far beyond its melting point in order to obtain sufiicient vapour pressure for opening the penetration passage and maintaining it open. This vapour pressure is increased with decrease in ray diameter and with increase in penetration depth since it has to counteract the surface tension of the molten metal surrounding the penetration passage, and the pressure of its liquid column.

Therefore, in most practical cases, it is necessary to select beam energy densities which excessively heat the material to be welded to a temperature far above the melting point and beyond the value required and sufficient for producting the weld connection. This excessive 3,448,240 Patented June 3, 1969 "Ice heating is generally greatest in the upper region of the weld seam so that, with known deep ray welding, the weld seam is V-shaped or wedge-shaped in cross section. To obtain the desirable seam configuration of a substantially rectangular or only slightly wedge-shaped weld seam cross section with known deep ray welding methods, it is necessary to select even higher beam energy densities so that the entire workpiece. In addition to the fact that a large part of this power is discharged uselessly from the side of the workpiece remote from the source of the beam, this causes a high loss of material to occur, and an uncontrollable gap may be formed, especially in the lower regions of the weld seam.

The excessive heating of the material necessary in the known method of deep ray welding, although less than that of other known welding methods, uses an unnecessarily high amount of energy. The accompanying phenomena occurring due to the penetration action when boring with the ray, more especially the material displacements from the welding region and within the welding region and the results of excessive material heating and evaporation, are even less desirable.

When the beam penetrates, the material is melted and until the beam has pierced the workpiece, the material is pushed out upwardly as a bead from the penetration region. When the beam pierces the workpiece, a material displacement occurs in the region of the outlet point of the beam in the direction of the beam which does not cause the material to well out, but does cause the molten material to spatter out of the weld seam. Due to the action of the high vapour pressure, this spattering occurs predominantly in the direction of the beam, but in the upper weld seam region, it also occurs in the opposite direction to the beam.

Thus, a material loss occurs which leads to shrinkage of the workpiece or, if the workpiece is unable to yield adequately, to the formation of cavities and, particularly at the root of the weld seam, to dangerous notches. These formations are caused by the rapid setting up of the material Welling from the seam region which prevents return of the material to the open seam as the beam moves across the workpiece. This explains why the production of faultless welds in accordance with known methods of deep beam welding is not sufficiently reliable. Only after an empirically found compromise of all welding parameters does this method lead to the required result in a limited field of applications, and in any case, is still very dependent upon minor variations in the welding conditions. Thus, the conditions to be maintained often require measures which are costly or not very practical.

Overheating the materials inevitably tends to form cavities and pores. Moreover, it influences the metallurgical processes in harmful manner. More especial-1y, it promotes a considerable change of the material composition and structure. A substantially uncontrollable temperature distribution occurs in the workpiece which usually doe-s not conform to the conditions required in ideal welding. The temperature distribution is predominantly dependent upon the boring operation and the energy absorption from the beam at the various boring depths. These conditions are quite different from the desired case of providing a controlled, definite energy distribution along the beam, for control of metallurgical processes.

It is, therefore, an object of the present invention to provide a method of welding workpieces by means of high energy 'beams and, more particularly, an electron beam, which is applicable with particularly advantageous results to welding workpieces of greater material thick-' ness than that previously possible with known deep welding methods, and which thereby eliminates the above described and other disadvantages of known deep beam welding techniques In accordance with the object, the method of the present invent-ion is briefly described as including the step of first forming an entrance or inlet opening at least at one point along the weld path. The opening is formed to a depth substantially equal to the desired depth of the weld seam. To initiate welding, the beam is positioned so that it is directed into the opening. The beam is adjusted so that at least a part of the wall surface of the opening :is heated to or beyond its melting temperature, the beam being then moved relative to the workpiece along the predetermined treatment path, the opening moving with the beam. Either the beam or the workpiece may be moved.

In the method according to the invention, the beam introduced in the opening has immediate access to the deep parts of the treatment path and need not first provide a penetration passage with the aid of vapour pressure.

In initiating the required treatment process, it is, therefore, necessary to supply only enough energy by means of the beam to cause the wall surfaces of the opening to melt. It has been proved that with the progressive movement of the beam, the melting region travels with the beam while maintaining its form. This may be explained in the following manner: If, for example, the opening is a bore which is produced mechanically in the treatment path, the introduction of the suitably adjusetd beam melts the material of the workpiece in the wall regions of the bore and forms a liquid layer on the material region which has not melted, and having a greater diameter than the bore. This liquid layer, due to its adhesion, retains the form of the unmelted material regions. If the beam is now progressively advanced in the treatment path provided, an increased heating of the wall region of the opening occurs in direction of travel of the beam, causing a thickening of the material layer melted at this point in comparison with the remaining wall reg-ions. This operation causes a change of the symmetry of the surface tension prevailing in the planes at right angles to the axis of the opening surrounded by the molten metal covering the walls of the 'hole. The forces released thereby force the molten metal in two flows extending symmetrically to the surface of the beam movement, in opposite directions around the beam towards the thinner layers in order to maintain the equality of the thickness of the layer wetting the wall of the opening, and hence maintain the balance of forces. In this way, the bore moves with the beam in the same direction and at the same speed through the workpiece.

In accordance with the invention, it is advisable to adapt the shape of the opening and the course of the beam to one another so that in each cross section taken at right angles to the direction of the beam, a portion of the beam output necessary for melting the walls to the required thickness is absorbed by the walls. In this way, a predetermined distribution of the beam energy may be ensured over the whole depth of the workpiece.

Furthermore, it is advisable for the beam parameters and the geometry of the opening to be so adapted to one another that the surface tension forces acting on the molten layer are in equilibrium with one another and with the adhesion forces when the beam stops at a given point of the processing path, but when the beam continues along the processing path, these forces become unbalanced and, therefore, force the molten metal located in the dircetion of movement and in front of the beam around and behind the beam.

Having briefly described this invention, it will be described in greater detail along with other objects and advantages in the following portions of the specification, which may best be understood by reference to the accompanying drawings, of which:

'FIG. 1 is a sectional view of a weld using the known method of deep beam welding;

FIG. 2 is a sectional view of a weld carried out in accordance with the method of the present invention;

FIGS. 3a, 3b, 3c, 3d and 3e are schematic plan views of two workpieces prepared for welding in accordance with the method of the invention;

FIG. 4 is a plan view of a modified embodiment of the method illustrated in FIG. 3;

FIG. 5 is a sectioned elevation view useful in explaining the processing of workpiece-s of irregular thickness in accordance with the method of the present invention;

FIGS. 6a and 6b are plan views illustrating two methods of carrying out a continuous welding process;

FIG. 7 is a. sectioned view illustrating a further embodiment of the present method;

FIGS. 8 and 9 are plan views illustrating still further embodiments;

FIG. 10 is a side view of a weld seam with wedge end pieces;

FIG. 11 is a plan view illustrating an alternative to FIG. 10;

FIGS. 12 to 18 are plan views illustrating various embodiments of the present invention;

FIG. 19 is a sectioned view illustrating an application of the method using a tilted beam;

FIG. 20 is a sectioned view along the line of the treatment path of a weld;

FIG. 21 is a sectioned view at right angles relative to the weld seam of the welding of two workpieces.

FIGS. 22 and 23 are partially sectioned views illustrating alternatives to FIG. 21 of the present invention; and

FIG. 24 is a sectioned view illustrating a multiple processing method.

In FIG. 1, there is shown a cross section through a weld seam 1 during its production in accordance with a known deep beam welding method. A typical process using an electron beam is disclosed in US. Patent 2,987,610. The beam 3 penetrating into the welding region overheats the central cross sectional region 2 so that the material evaporates and forms a tapered bore. The overheated region extends over the entire central cross sectional region and has a width which decreases with depth of penetration so that the region 2 tapers to a point. Region 2 is wedge shaped to a greater or lesser extent depending upon the amount of the excess power applies and on the form of the beam. So long as the beam 3 has not penetrated the workpiece, the molten material is formed out upwardly from the penetration region as a bead 4, and also flung vertically upwards and outward due to the creation of a high vapour pressure in the hole. As soon as the beam 3 penetrates the workpiece, a movement of molten material is initiated vertically downward, which not only results in material Welling out at the point 5, but also in ejection of molten material in the form of particles 6. The material loss occurring in this manner may lead to shrinkage and the formation of cavities. The considerable overheating of the mate rial also produces a tendency to pore formation and may distort or harmfully afiect the structure of the material.

In FIG. 2, there is shown the application of the present invention to the same kind of work. The workpiece 10 is first provided with a cylindrical bore 11 (size exaggerated for clarity). An electric charge carrying beam 12 is so focused against the surface of the workpiece 10 that it converges and diverges in the bore and hence irradiates the said wall surfaces. These wall surfaces become heated and assume the form indicated by the broken line 13. The dotted line 14 is the boundary between the liquid and the solid material.

The forces resulting from surface tension, cohesion and adhesion retain the molten layer in the position shown as long as the thickness of the layer is not excessively enlarged by the energy supply from the beam. If the beam energy is reduced or switched oil, then the temperature of the heated material region drops and the molten material solidifies in substantially the form as shown. The

molten layer

13, 14 shows no tendency whatsoever to close the path enlarged through the bore by the beam in a manner which would substantially impair the heating process. Therefore, no further energy from the beam is required to keep open the bore 11 such as is required in the known method of deep beam welding to evaporate the material. If the beam is progressively displaced, then the opening drifts with the beam. As the beam is displaced, the material melted by the ray will increase in the front of the ray. However, surface tention causes flow of material along the sides of the bore to equalize the layer depth around the bore. Thus, as the ray progresses, a weld is formed behind the ray.

In known method of deep beam welding, as shown in FIG. 1, the resulting inwardly acting forces of surface tension are not cancelled by adhesion since the molten mass Welling out of the workpiece and subject predominantly to the influence of surface tension strives to return to the hole and, thus, renders it necessary to produce equivalent counterforces which are supplied by increasing the intensity and heating action of the beam. This circumstance necessitates the overheating initially referred to, with its corresponding disadvantages. The same applies to the welding movement relate to the workpiece. Herein also a substantially greater expenditure of energy is required according to the known method shown in FIG. 1 than in the method in accordance with the invention shown in FIG. 2.

In FIG. 3, there is shown two

steel plates

20 and 21 which are to be welded along the abutting edges 22. In order to prevent irregularitie from occuring in the weld seam caused by the inlet opening required at the beginning of the seam and the opening remaining at the end of the seam, two

blocks

23 and 24 of equal thickness to the workpiece at the separating line 22 are placed close to the work at the beginning and end of the line, as shown in FIG. 3a. The block 23 has an inlet opening in the form of bore 25 formed therein. This bore 25 acts as the starting point of the Welding, which is carried out right into block 24. The

blocks

23 and 25 are subsequently removed.

The inlet opening may also be provided at a point located in the intended weld seam 22. FIGS. 3b and 30 indicate some further possibilities of placing the inlet opening directly at the beginning of the separating line 22. Block 23 is of different form in each figure. 4

Since with the present welding method, substantial reduction of energy expenditure is possible, the usual results of welding occur in a corresponding reduced manner. The usual disadvantages arising from the material becoming overheated in the old method of deep beam welding are avoided, but a certain amount of shrinkage, for example, is experienced. This may cause a reduction of the original aperture size to occur during welding of a long seam. In these cases, it is advisable to provide additional openings along theweld seam which are reached by the welding beam before the traversing aperture has been appreciably reduced, so that the opening is widened again to its original extent. In FIG. 4, which corresponds to FIG. 3, such openings are shown. The opening 26 is a bore formed symmetrically to the weld line and formed as two grooves of optional profile in the

plates

20 and 21. At 27, a groove is shown formed in one side of plate 20 which also fulfills the required purpose.

With continuous or intermittent thickness variations of the workpieces along the weld seam, it is possible to ensure that free access of the beam energy in depth is maintained by providing a succession of bores or grooves along the treatment path. An example is shown in FIG. 5.

This figure shows a section along the weld path and illustrates the method of compensation for different variations (steep and gradual) in thickness of workpiece 30. If welding is to be carried out from left to right, a block 31 in accordance with the method shown in FIGS. 3

and 4 is first mounted in position having the same thickness as the left hand end section 38 of the workpiece with an opening 32 for the entry of the beam. Welding is then carried out in the same manner as in FIG. 3 progressively from this opening into the actual weld seam. The opening 32 moves progressively with the ray out of the block 31 into the workpiece 30. As soon as the beam and the opening moving therewith reach the left hand side of the thicker workpiece section 39, the beam power output suitable for the thickness of the left starting section 38 no longer suflices to keep the wall of the opening fluid throughout the thicker section 39. Therefore, when the beam enters this thicker section, it is necessary to ensure that the beam output available per unit area of the opening remains at least substantially constant. This may be attained with the opening diameter remaining unchanged by correspondingly increasing the beam output or, when the beam output remains constant, by correspondingly reducing the bore cross section. It is also possible for both the diameter of the hole and the beam output to be varied so that the proportion of beam output available per unit area of wall surface of the hole remains substantially unchanged. To avoid having to form an opening at the left hand end of the thickened section, a block 33 having a thickness equal to the thickness increase is mounted at the starting section of the thickened region 39 as shown in FIG. 5, and which has an opening 34 formed therein in a manner corresponding to FIGS. 3 and 4. The beam is guided in the workpiece section 38 to the region of the bore 34 of the block 33 and then, if necessary, adapted to the depth of hole of the thickened region 39, by increasing its intensity, its impulse duration (in impulse controlled rays), its focusing state or the like so that the walls of the bore 34 and the beam bore (not shown) above it are heated and liquefied by the beam in the required manner. Then the beam is continued in the treatment path to the right in FIG. 5 and, when reaching the workpiece region 40 where the workpiece thickness is continually reduced, adapted continuously to the reducing thickness of the workpiece until at point 41 it has substantially the same parameters as in the starting section 3-8. Here, the thickness of the workpiece increases again, but gradually, so that the use of a block in the manner of

blocks

23 and 31 is difficult. In this case, the action is that the beam intensity is progressively increased to correspond with the workpiece thickness, additionally formed holes such as 35 and 36 shown, supply the beam bore with new volume, so that the diameter of the beam bore does not drop below a minimum value. The

holes

35 and 36 thus have to fulfil the same object as the

additional openings

25 and 26 in FIG. 4.

Finally, the end section 30 of the workpiece is traversed with the beam parameters remaining constant, and the beam is passed through the mounted block 37 out of the region of the workpiece 30. The block 37 thus has the same function as block 24 in FIGS. 3 and 4. To com pensate for shrinkages or thickness increases, it is possible to begin with an opening having a relatively large diameter' and then with the progress of the welding operation to accept the consequent reduction of the opening diameter. In this case, beam output and beam diameter are so adapted to the momentarily occurring diameter that the beam output per unit area of the opening remains substantially constant.

FIG. 6 shows welding of two workpieces 43' and 44 along an endless line 45. In this case, if no opening must remain in the line 45 itself, the opening in which the beam performs its work after line 45 has been traversed is led to a point lo cated outside the line. The end of the treatment path is in this case located in a mounted block according to the example shown in FIGS. 3 and 4. FIG. 6411 indicates such a block 46 and the broken line 47 indicates how the beam together with the bore 48 carried along with it can be led out of the line 45. It is, of course, also possible for the beginning of the treatment path to be located in such a mounted block, and it is even possible for one and the same block to be used for the beginning and the end of the treatment path. In some cases, it may be undesirable for the entry or exit path 47 to be as symmetrically disposed in the workpiece. It may, for example, be desirable for the thermal and mechanical stresses caused by the passage of the beam in the entry or exit path 47 to be made symmetrical to the workpiece axis. In such cases, it is also possible to provide several entry and exit paths in symmetrical distribution, as shown in FIG. 6b which has two opposite entry or

exit paths

47a, 47b, aligned radially relative to one another. The beam is then guided first from the opening 48a in

path

47a, 45a, 47b, to the block 46b and then from block 46b in the

path

47b, 45b, 47a back into block 46a.

The effect that the opening or bore occupied by the beam contracts to a smaller diameter with an increase of the workpiece thickness may be used in cases in which the bore is to disappear within the workpiece.

As shown in FIG. 7, wedge-like blocks 49 and 50 placed in position simulate a gradual increase of the workpiece thickness, so that the bore led out of the starting block 51 and into the workpiece in a manner described with reference to FIGS. 3 and 4, becomes progressively smaller and finally closes in the region of the wedge-shaped blocks 49 and 50. In this case, when the beam enters the region 53, it may be expedient to increase the beam output in accordance with the increase of the material thickness and/ or to reduce the beam diameter in order to maintain the wall surface of the bore at the required temperature over its whole length.

FIG. 8 shows a method of carrying out the present process in which two

workpieces

54 and 55 to be welded together are not supported against one another at their connecting line provided, but are separated by a gap. This case may be regarded as a border line case of the method explained in FIG. 4. In accordance with this, gap 56 corresponds to a large number of

additional bores

25, 26. In this case, the bore into which the beam moves must be large enough for the feed of the liquefied material running around the beam to provide the material for filling the gap at the end of the weld.

The choice of a method in accordance with FIG. 4 or 8 depends mainly upon the material properties, more especially upon the shrinkage occuring during welding. In some cases, a gap 56 may be expedient because even in considerable workpiece thicknesses, a required distribution of the beam energy is more readily obtainable over the depth of the weld seam. For this purpose, the mass distribution in the gap may be preformed in accordance with the depth. This may be realized in various ways. For example, as shown in FIG. 9, it is possible to insert a net or pins in a gap of uniform Width, the distribution of the mesh wires of the net or the pins being so arranged that in the individual regions of the gap depth, predetermined proportions of the beam output are absorbed by the wire mesh or pins and transmitted to the adjacent regions of the gap wall surfaces. In this way, it is possible even with relatively low energy beams to obtain a required distribution of the beam output over the depth of the weld seam.

If on welding while employing a gap 56, there is a risk of the material flowing into the gap causing a notch to be formed in the weld seam (normally this disk does not occur as a result of the shrinkage observed), additional material may be supplied, for example, by means of Wedges 58 as shown in FIG. inserted in the gap.

A further modification is shown in FIG. 11 where one wall surface 59 of the gap 56- is undulated, fluted, or provided with raised portions. Such surface forms may provide advantageous features. In materials which, during welding, have a considerable tendency to shrinkage, this may be counteracted. The distribution of the beam output over the depth of the gap may be controlled in a manner similar to the mesh referred to above, and with a given maximum width of gap (which may be required for reasons of beam geometry), the average gap may be adjusted to a smaller value, which could be required to prevent sagging weld seams. The use of a gap is particularly expedient when welding inaccurately machined surfaces, where the variations in spacing between the surfaces to be welded are accordingly reduced along the treatment path.

FIG. 12 shows the possibility of employing gravitational force to assist in the Welding. The treatment path or intended Weld seam 60 is vertically arranged and welding carried out by moving the beam upwards, so that the material melted at the beam front is forced by gravity into the gap closing behind the beam. The

blocks

61 and 62 shown correspond to the

blocks

23 and 24 of FIG. 3.

Instead of the force of gravity, or additionally thereto, other forces may be employed to assist the material feed occuring around the ray in the opposite direction to the beam movement. Thus, FIG. 13 shows the possibility of employing an electric current 63 which flows in a direction as indicated by the arrows 64 around the given weld position 65. The beam is displaced in the direction opposite to that indicated by the arrow 66. The maximum density of the electric current 63 occurs at the point 67 where the molten material, due to the effect of its own magnetic field, is subjected to a force acting in the direction 66 and resultant from the magnetic fields produced by the

current conductor sections

68 and 69. It is evident that also by other methods an electrodynamic interaction can be produced between the molten material and a magnetic field, to exert a repelling force on the portion of the molten layer located in front of the beam.

In FIG. 14, the double arrow 70 indicates that additional forces may be exerted on the material melted in the welding region by means of a vibration of the workpiece effected in the direction of movement of the beam. [t is obvious that due to a non-sinusoidal vibration, more especially in conjunction with a beam reciprocated additionally in the direction of movement of the ray, forces of inertia may be produced in the liquefied material which assist the displacement of the material to the rear of the beam. Such effects may also be produced or amplified by means of an impulse control of the beam synchronous with the vibration. In some cases, it may be expedient to produce a vibration at right angles to the treatment path.

FIG. 15 shows the possibility of assisting the displacement of the liquid material by centrifugal forces. The workpieces 72 to be processed along the line 71 are arranged on a carriage 74 which is displaceably mounted on a turntable 75 in the direction of the arrow 73, at right angles to the direction of the beam 76. The turntable is rotated about the axis 77 in the direction of the arrow 78, and the beam is moved around in a circle synchronously therewith so that the beam impact point 76 does not rotate relative to the workpieces 72. By rotating the workpieces, a centrifugal force is exerted at the beam impact point 76 against the material liquefied there, the force being proportioned to the distance of this point from the axis 77 and the speed of the turntable 75. By varying the radius of rotation of the beam and thus the distance between the beam impact point 76 and the pivotal axis 77, the centrifugal force may be varied. By displacing the carriage 74 on the turntable 75, the processing point 76 is displaced on the treatment path parallel to the direction of movement of the carriage, the centrifugal force exerted remaining constant, unless the radius of rotation of the beam is varied. FIG. 15 shows that welding is carried out in the direction of the axis 77 so that the centrifugal force acts on the liquefied material in a direction opposite to the direction of movement of the beam, thus, assisting material movement caused by the surface tension.

FIG. 16 shows how the surface tension of the material layer melted by the beam may be reduced in the treatment path, in this case a weld gap, in the manner of the gap shown in FIG. 8 by the introduction of a surface tension reducing agent 79. It is expedient to ensure that the concentration of the additive is reduced rearwardly and preferably becomes zero behind the beam. This may be readily obtained by the use of evaporating additives.

Such additives are known, for example, a flux as used in conventional welding technique. Reduction of the surface tension may be desirable especially when operating with very fine openings or bores, when a particularly narrow weld seam is required. In this case, the use of additional forces in the manner as described with reference to FIGS. 12-l5 may be particularly expedient. Additives may also be brought to the welding point in the form of a wire by blowing powder thereon or by injecting a material jet comprising the additive material.

FIG. 17 illustrates methods of carrying out the invention particularly useful with relatively large openings or bores. In FIG. 17:, a beam is used, the cross section of which is considerably smaller than the cross section of the opening 80, and the beam is circulated rapidly around the edge of the opening. In this way, the beam only acts on the edge regions of the opening 80 and comes fully into effect only for melting these edge regions. More especially, in the case of electron beams which are readily controllable with known aids and practically without inertia, this method permits relatively large openings to be irradiated with sharply focused beams in such a manner that the wall of the opening is melted in the required manner. Relatively large openings may be irradiated also with a beam, the intensity of which is preferably concentrated in a ring region instead of with a rotating beam. The production of such hollow beams is known in the relevant art. Hollow beams, having an asymmetric intensity distribution may be produced by faults in the optical system of the beam-producing means. Hollow beams may also be produced by artifically invoked astigmatic or coma errors, and may result from tilting the lens system relative to the beam axis. FIG. 17b shows in an illustration corresponding to FIG. 1711 the irradiation of the edge region of a relatively large opening 81a by the annular intensity 80a of ahollow beam.

A particular advantage of the method shown in FIG. 17 of the beam passed along the wall surface of the opening is that openings having a noncircular cross section, for example, openings having an elliptical cross section, may be used since it is readily possible to move the beam in a noncircular closed path. Moreover, during each circulation of the beam, its parameters may be controlled in a predetermined manner. Thus, when passing through the part of the closed path of circulation ahead of the direction of beam movement, the intensity of the beam may be somewhat increased and/or its speed of rotation may be somewhat reduced in order at this point to obtain increased melting of the workpiece material. Conversely, increased melting may be obtained in-the part of the path of circulation or the periphery of the opening located behind the direction of the beam. By suitable control of the beam parameters during the rotational movement of the beam, a bore or opening which at first has a circular cross section may be reshaped to have an elliptical or any other cross sectional form. An elliptical cross section shape presents certain advantages with regard to the mechanism of the material drift, due to the concentration of the surface tension forces in two opposite wall regions. These advantages are more clearly demonstrated if instead of an elliptical, a substantially egg-shaped cross section is used.

FIG. 18 illustrates such a case and shows in plan view a weld seam 82 between two workpieces 86. The opening 83 moving progressively with the beam in the direction of the arrow has a substantially egg-shaped cross section owing to the beam control selected. In the wall region 85 located in the direction of the beam, the radius of curvature is greater than in the wall region 84 located behind the beam. The result of this is that the material melted by the beam in the front edge region is assisted in its tendency to flow to the rear wall region 84.

This effect may be increased by increasing the beam output applied to the front leading edge region, for example, by slowing down the speed of rotation of the beam and/or increasing the beam output. Each measure which leads to a more rapid melting of the front wall region and to a more rapid conveyance of the material melted there to the rear, is advantageous in increasing the processing speed.

The beam may have a linear additional movement imparted to it, such as a reciprocating movement extending in the direction of the welding operation, or a swinging movement executed at right angles to the weld seam. In this case, also by controlling the beam movement and/or the beam parameters, a momentarily modified distribution of the beam energy may be obtained against the Wall surface of the bore and/or its regions.

Further advantageous effects may be obtained by peri odically altering the focus of the beam, or periodically tilting it about the bore axis. In this way, it is possible to obtain any required distribution of the beam output over the different depth regions of the bore. These measures may be applied both on their own and in combination with the remaining possibilities of beam control, such as guiding the beam in circular or elliptical paths, impulse control and so on.

FIG. 19 shows in a sectional view along the treatment seam a possible pivotal movement or tilting of the beam. The opening 88, shown exaggeratedly large in the welding plane 87 is irradiated by a thin energy beam which is periodically tilted about the opening axis 91 and simultaneously rotated rapidly about this axis. With a relatively large tilting angle 89, the upper wall regions of the opening 88 are irradiated, and with a small tilting angle 90, the lower wall regions are irradiated. It is obvious that similar effects may also be obtained by varying the focus of the beam.

FIG. 20 shows the possibility of controlling the beam so as to vary it automatically in accordance with varying workpiece thicknesses. Below the treatment path of two workpieces of irregular thickness 92 to be welded, a collector finger 94 is mounted below the beam entry point 93, which receives the beam portion emerging downwardly from the treatment path and produces a control signal which is a measure for this portion. Collector 94 and beam generator are both simultaneously moved relative to the workpieces 92. The control signal acts as control of the beam generator via a control coupling 96 as by interposition of a control device 97 which converts the signals supplied by the collector 94 into control valves for the beam generator 95. The beam generator is preferably so controlled that the proportion of beam energy received by the collector remains substantially constant. This regulation automatically increases the beam intensity to the correct amount if the beam moves in a region 98 of increasing workpiece thickness, and vice versa.

The regulating device shown in FIG. 20 may also be used to regulate the rate of movement of the beam in the direction of treatment so that the portion of beam energy received by the collector 94 remains substantially constant.

A signal giving information concerning the state of the work at the point of treatment may also be obtained with other measuring feelers, as by means of temperature or beam measuring devices, mechanical feelers which are guided in slots or grooves parallel to the treatment path or in the treatment path itself, or other devices which are moved with the beam along the treatment path. Signals supplied from these or other sensing devices may provide control values for the beam, for example, for trailing the beam, intensity control, etc. or for the treatment operation to be otherwise modified so that an automatic control substantially in the manner as explained with reference to FIG. 20 may be obtained. Several measuring devices for different values or states may be used simultaneously. The present method is advantageous in many instances where a hidden weld seam or one difiicult of access is to be made. An example thereof is explained in the sectional view of FIG. 21. The two

workpieces

99 and 100 are to be connected at the weld seams 101 and 102 to form a T- section. This may be effected in acordance with the method of the invention by guiding the treatment beam 103 with its bore 104 in a path located over the required weld seams 101 and 102 by means of the cross piece 99. Welding then takes place due to the beam portion emitted at the lower end of the bore 104. Behind the beam, the material closes in again to form an endless surface. The action may be started with the aid of a block 105 mounted on the cross piece 99 and having an opening formed therein, as described with reference to FIGS. 3 and 4. The other described measures, for example, automatic beam control as shown in FIG. 20 may be employed.

The present method also permits T-sections to be butt welded. This is shown in FIG. 22. Two

sections

106 and 107 of fiat material are arranged at right angles to one another along the butt joint 108. In one workpiece 107, the entry opening 109 is formed at the beginning of the butt edge 108. The beam 110 is introduced from above into the opening 109 and then guided along the butt edge 108. In this case, the beam or aperture diameter is selected to be substantially equal to the thickness of the workpiece 106 so that welding takes place at the lower end of the opening 109 over the whole width of the joint butt edge. If, as shown, the preformed inlet opening 109 is made as a notch on the left edge of the workpiece 107, it is possible to mount a block 111 laterally in a similar manner to that shown in FIGS. 3 to 8, which in the case shown, consists of a piece of fiat material. It is obvious that this piece 111 may also be welded along the butt edge 112 with the piece 106. Accordingly, a second block 112 is provided at the end which, if required, may be welded to the piece 106.

FIG. 23 shows an alternative to FIG. 22 for producing a T-section from two

workpieces

114 and 115 welded along the butt edge 113. The inlet opening 116 in this case is formed in the laterally mounted block 117. At the end of the joint or weld seam 113, a second block 118 is mounted which receives the opening remaining at the end of the welding process. The

blocks

117 and 118 have substantially the same thickness as the piece 114.

FIG. 24 shows three simultaneous treatment procedures. As shown in FIG. 24, three

sheets

119, 120, 121 shown in section are to be welded in the sectional plane of the figure. The three sheets are arranged as shown with superposed

treatment paths

129, 130, 131 and starting

edges

132, 133, 134. At the starting edges, blocks 123, 126, 128 are mounted which have

inlet openings

122, 125, 127 formed in alignment therein. The beam 124 is introduced into the

openings

122, 125, 127 and passes through all three openings successively. In this case, the opening diameters and/or the beam course may be so selected that each of the three openings receives substantially the same proportion (or, if the

workpieces

119, 120, 121 vary in thickness, a corresponding proportion) of beam output. The treatment as described, is then carried out in accordance with FIG. 24 from left to right and terminates in the

blocks

135, 136 and 137.

As evident from the above statements, the present method differs fundamentally from known methods of deep beam welding with regard to welding. The workpiece positions to be treated are so prepared by suitable shaping that at the starting point of the treatment path, the energy beam has free access from the start to the material to be melted right throughout its depth. It is essential that the beam on entering the opening prepared for it, so irradiates the walls thereof that the beam, after a period depending upon its intensity and the material and workpiece properties, melts the walls substantially completely to a thickness required for welding. The distribution of the layer of molten material resulting from the arrangement and shape of the opening formed in the workpiece, and the heat action exerted by the beam against the wall surfaces, is decided by the further course of the welding operation. The thickness and distribution of this layer are preferably so determined that when the beam is stationary relative to the workpiece, all forces acting on the molten layer caused by the surface tension are in equilibrium. On the other hand, if the beam is displaced relative to the workpiece, the equilibrium of the forces is disturbed so as to urge the molten material located in front of the beam in the direction of movement around to a position behind the beam, so that welding takes place behind the beam.

The said method may be carried out with a variety of types of energy beams, for example, light beams, laser beams, electron beams and the like. In practice, electron beams have proved particularly successful since they are readily controllable and are without inertia and permit high output feeds. The beam power used depends upon the kind and thickness of the material to be treated, the required processing rate and other data differing from case to case. For normal workpieces in engineering, electron beam generators having beam outputs up to some kilowatts suffice, e.g., beam generators having 150 kv. acceleration voltage and 50 ma. beam current. Known techniques of impulse control may also be employed in the said method.

EXAMPLES (A) Conventional electron beam deep welding Two pieces of chrome nickel steel specification AISI 302 (corresponding to German specification X 10 CrNi 189) of 54 millimeters thickness were buttwelded together along a seam of 10 centimeters length. The beam acceleration voltage was kilovolts, and for the selected welding velocity of 2 millimeters per second it was necessary to adjust the beam current to 50 milliamps (corresponding to a beam power of 7 kilowatts) to obtain the desired full welding depth of 54 millimeters. After welding, the structure of the weld seam was inspected at an etched cross-sectional cut taken in the last third of the seam length. The weld seam cross-section was generally wedge-shaped. On the workpiece side facing the beam generator, the material was welled out to a height of about 5 millimeters in a zone of about 12 millimeters width. Below this entrance end of the seam depth, the material had been fused along a groove of approximately semicircular cross-section and about 5 millimeters depth. Therebelow, the weld seam continued from the innermost range of the groove to the opposite surface of the workpieces, beginning with a width of about 5 millimeters and ending with a width below 1 millimeter.

(B) Novel method Two pieces of the same composition and dimensions as in Example A were in like "manner welded together along a seam of 10 centimeters length. The same beam generator as in Example A was used with the focussing condition and acceleration voltage unchanged. According to FIG. 3a, a block having a bore of 0.6 millimeter diameter was placed at the one end of the butt. The beam was introduced into this bore, and the welding was performed in the manner disclosed with the bore moving together with the beam. At the same welding velocity of 2 millimeters per second as in Example A, only a beam current of 35 milliamps (corresponding to a beam power of 4.9 kilowatts) was necessary to obtain the full desired Welding depth of 54 millimeters.

As in Example A, the structure of the weld seam was inspected at an etched cross-sectional cut taken in the last third of the seam length. The weld seam was generally rectangular in cross-section. The width of the seam varied over the depth between 1.5 and 3 millimeters and was smallest in the first third of the seam depth. No material was welled out.

These examples show that the novel method leads to narrower weld seams of more uniform width with considerably lower energy requirements.

The invention may be variously modified and embodied within the scope of the subjoined claims.

I claim:

1. The method of thermal welding of workpieces by a high energy ray which comprises the steps of forming an entrance aperture into said material to a depth substantially equal to the desired depth of the weld seam, directing the beam into said preformed entrance aperture to heat the wall of said opening and to melt only a thin layer of said wall so that the layer of said melted material will adhere to said wall despite the molten condition and resist the gravitational pull urging said material downwardly, and moving said beam along the desired path of treatment at a speed which is maintained sufliciently low so as to permit the material melted on the wall in advance of the beam to flow and equalize on the wall behind the beam to close the aperture behind the beam to provide a weld seam as the beam is moved over said predetermined path.

2. The process of claim 1 wherein said entrance opening is a through bore.

3. The method of claim 1 in which a block provided with said entrance opening is placed close against said workpiece at the starting point of the desired treatment, said treatment path including the distance between said entrance opening and said starting point, whereby said ray and said opening travelling therewith are guided out of said block and into said workpiece at the beginning of their travel along said treatment path.

4. The method of claim 1 in which a block is placed close against said workpiece at an ending point where said processing of said workpiece is to end, said treatment path including the distance between said ending point and a point on said block whereby saidray and said opening travelling therewith are guided out of said workpiece and into said block at the end of their travel along said treatment path.

5. The method of claim 1 wherein the parameters of said ray are adapted to the thickness of said workpiece during transverse of said treatment path.

6. The method of claim 5 wherein at points of said traetment path where a step-like increase of the thickness of said workpiece occurs on the side of the workpiece remote from said ray, a block of the same thickness as said steplike increase and having a bore of substantially the same dimensions as the opening moving with said ray is placed close to the step so that said bore is arranged below said treatment path, said ray and said opening moving therewith coming into coincidence with said bore during their movement along said treatment path.

7. The method of claim 1 in which additional openings are provided in said treatment path to compensate for decrease in size of the opening in such cases where the free volume of said opening moving with said ray tends to gradually decrease such as with workpieces which have an increasing thickness over at least a part of the length of said treatment path or tend to shrink during the treatment, the dimensions and spacings of said openings being chosen so that said decrease of the free volume of said opening moving with said ray is at least approximately compensated.

8. The method of claim 1 wherein in such cases where the free volume of said opening moving with said my tends to gradually increase such as with workpieces which have a decreasing thickness over at least a part of said treatment path or tend to grow during the treatment, additional material is supplied during said processing in such quantities that said increase of said free volume is at least approximately compensated.

9. A method according to claim 1, wherein .a gap is provided between at least two surfaces of workpieces to be welded together, and wherein additionl material is located in said gap in such quantity and distribution that predetermined portions of the total power supplied by said ray are absorbed and transmitted to said surface in the various depth regions of said gap.

10. The method of claim 9, wherein said additional material is provided in the form of a mesh or grid in the gap.

11. A method according to claim 1, wherein a gap is provided between at least two surfaces of workpieces to be welded together, and wherein at least one of said surfaces is contoured so that predetermined portions of the total power supplied by said ray are absorbed in the various depth regions of said gap.

12. The method according to claim 9, wherein said additional material includes at least one substance capable of changing the surface tension of the material melted in said gap by action of said ray.

13. The method of claim 1, wherein an additional force is exerted on the material liquefied by the action of said ray.

14. The method of claim 13 wherein said treatment path is arranged vertically and is transversed upwardly by said ray whereby the force of gravitation becomes effective as an additional force to facilitate the back-flow of the material melted at the front of the moving ray.

15. The method of claim 13 wherein the workpiece material is subjected to vibratory movements in such manner that the material melted at the front of the moving ray is subjected to inertia forces tending to move said melted material in a direction opposite to the direction of movement of said ray.

16. The method of claim 13 wherein said additional force comprises a centrifugal force.

17. The method of claim 16 wherein said workpiece is rotated with the center of rotation being spaced from the point of treatment in the direction of movement of said ray in said treatment path whereby a centrifugal force is created which tends to move material melted at the front of the moving ray in the direction opposite to said direction of movement of said ray.

18. The method of claim 15 wherein said vibratory movement is nonsinusoidal so that the wall region laying at the front of said ray in said opening moving with said ray receives an increased portion of the power supplied by said ray whereby melting of said wall region is favored.

19. The method of claim 15 wherein the ray parameters are changed in synchronism with said vibratory movement so that the wall region lying ahead of said ray in said opening moving with said ray receives extra energy to facilitate melting thereof.

20. The method of claim 13 wherein said additional force is generated by interaction between a magnetic field and an electric current flowing through the molten material at the point of treatment.

21. The method of claim 20 wherein said electric current is induced by means of said magnetic field.

22. The method of claim 20 wherein said magnetic field is generated by said electric current.

23. The method of claim 1 in which said ray is, in addition to its movement along said treatment path, moved relative to said opening in the area where said ray impinges on said workpiece.

24. The method of claim 23, wherein said ray has a smaller cross section than said opening and is circulated around said opening, the shape of the circulatory path corresponding to that of the edge of said opening.

25. The method of claim 24, wherein said circulatory path is egg-shaped, the portion having the smallest radius of curvature pointing in the direction opposite to the direcgon of movement of said ray along said treatment pat 26. The method of claim 23, wherein said ray is pe riodically tilted about its mean axis.

27. The method of claim 1, wherein at least one parameter of said ray is controlled by signals derived from feeler means responding to performance data of said processing or changes of state caused thereby.

28. The method of claim 27, wherein said ray and said feeler means are moving together along said treatment path, said feeler means being arranged in the proximity of said treatment path.

29. The method of claim 27 in which said opening is going through the full thickness of said workpiece, in which at least one measuring feeler is provided on the side of said workpiece remote from said ray to receive that portion of the ray which had passed through said opening, and in which the parameters of said ray are controlled so that a signal supplied by said measuring feeler remains substantially constant.

30. The method of claim 1 in which additional substances are fed to the treatment point.

31. The method of claim 30 in which said additional substances are fed in the form of a beam of material particles.

32. The method of claim 1 in Which a wedge increasing in thickness in the direction of the movement of said ray is placed close to the point of said treatment path where said processing is to end without leaving an opening, said ray being guided into the part of the treatment path covered by said wedge with a substantially constant power output so that upon attaining a certain overall thickness of said workpiece and wedge said opening closes.

References Cited UNITED STATES PATENTS 2,987,610 6/1961 Steigerwald. 3,230,339 1/1966 Opitz et a1. 3,258,576 6/ 1966 Schleich et a1. 3,291,959 12/1966 Schleich et a1.

OTHER REFERENCES RICHARD M. WOOD, Primary Examiner.

W. DEXTER BROOKS, Assistant Examiner.

US. Cl. X.R. 219-117