CROSS-REFERENCE TO RELATED APPLICATIONS

The present application is a 35 U.S.C. Â§Â§371 national phase conversion of PCT/EP2013/073711, filed Nov. 13, 2013, which claims priority of European Application No. 13153575.9, filed Feb. 1, 2013, the contents of which are incorporated by reference herein. The PCT International Application was published in the German language.

TECHNICAL FIELD

The invention relates to a welding method, in which the welding power is adapted to the thickness of a substrate.

TECHNICAL BACKGROUND

In the case of components which are subjected to loading and are produced from nickel-based superalloys solidified in polycrystalline form, marginal-layer remelting of cracks close to the surface by means of laser radiation is desired in order to close the cracks and in order to retain the mechanical properties of the components to be repaired in the region of the base material. The remelting of regions of a turbine blade or vane having relatively large variations in the material thicknesses (3 mm-1 mm) can lead to different molten bath depths when remelting with a constant laser power at the molten bath surface.

SUMMARY OF THE INVENTION

It is therefore an object of the invention to solve the aforementioned problem.

Investigations carried out to date on sample geometries of nickel-based superalloys show that the deviations in the remelting depth are reduced by 50% compared to remelting with a constant laser power as a result of remelting with a constant process temperature at the molten bath surface.

However, an advantageous remelting depth along the remelted path is achieved only by a path-dependent welding power calculated in advance (a priori).

BRIEF DESCRIPTION OF THE DRAWINGS

FIGS. 1, 2 and 3 show procedures or results, according to the invention, of the method,

FIG. 4 shows a list of superalloys.

DESCRIPTION OF EMBODIMENTS

The figures and the description represent only exemplary embodiments of the invention.

FIG. 1 schematically shows a component 1 comprising of a substrate 5 which has been subjected to a welding method on the surface 14. The welding power P is shown as a function of the thickness dâ€², dâ€³ along the section x or the time t.

In a region 4, the substrate 5 has a thinner wall thickness dâ€³ (<dâ€²), because a duct is present internally there, for example.

This is the case, for example, in turbine blades or vanes which have internal meandering cooling ducts delimited by walls. A cooling duct of this type represents a recess or a thinner wall region 4.

The substrate 5 is preferably to be remelted at the surface 14, or build-up welding is to take place. The method hereof is particularly suitable for remelting.

There is no regulation of the welding power on account of the temperature of the molten bath surface in the region of the focus of the welding beam 16 (FIG. 3). Instead, a simulation of the molten bath depth is carried out a priori on the basis of heat conduction calculations along the welding path 13 and welding direction 19.

The aim of adapting the welding power P is preferably to achieve a constant thickness of the molten or applied region 13.

This is shown in FIG. 1, in which a certain welding power P is present at the start 8 in the thicker region dâ€². This welding power is already being reduced 9 when the welding beam 16 (FIG. 3) approaches the thinner (dâ€³) region 4 or has not yet reached it.

The welding power P reaches a minimal value 15 only when the region 4 has already been reached by the welding beam 16. The welding power P also increases further after the thinner (dâ€³) region 4 has been crossed, until it reaches a higher value 7 again.

The wall thickness dâ€², dâ€³ is the thickness in the direction of a welding beam 16 (FIG. 3) or in the direction of build-up of the material during the build-up welding, but not parallel to the surface 14.

The curve profile of the welding power P is shown only schematically.

It is significant that the welding power P is not configured in a manner corresponding to the contour of the region 4, i.e. it is not reduced in a ramp-like manner, but instead is lowered in advance before the thinner (dâ€³) region 4 is reached and also only rises slowly again preferably after the thinner (dâ€³) region 4 has been left, and is not increased again in a ramp-like manner.

This is caused by the heat conduction, which has the effect that regions still to be remelted are also heated in advance already by a flow of heat and the introduction of energy required for melting and heating is reduced. The heat conduction is influenced by the introduction of heat by the welding beam 16 and the cross section (thickness) of the region 13 to be melted and also the thickness of the material transverse to the direction of movement 19 of the welding beam 16.

FIG. 2 shows a test result, in which a region 4 having a thinner wall is present, and a remelting region 10, having a relatively constant remelting depth.

In FIG. 1, it has been assumed that a maximum crack depth is present, on the basis of which the thickness of the remelting region 13 which is to be remelted is stipulated.

It is also possible, however, for the crack depths to be greatly different along a direction of movement 19, and, in the thicker region dâ€², to exceed the thickness dâ€³ of a thinner region (FIG. 3).

Here, too, it is possible to calculate the welding power in order to remelt the crack in an optimally and locally adapted manner at its different crack depths, in order to achieve remelting regions 10â€² of differing thickness or thicker remelting regions 10â€².

The method is preferably carried out for nickel-based or cobalt-based superalloys.

The substrate 5 is preferably a nickel-based or cobalt-based superalloy, also very particularly as shown in FIG. 4.