To provide tailored, lightweight and cost-efficient components, joining dissimilar metals with superior quality is important. Laser welding continues to show great potential for such applications, enabling high-precision joining with low heat input and the customised mixing of dissimilar metals.
A promising approach to further improving the laser welding of dissimilar metals is the introduction of ultrasonic excitation into the weld pool, which could enable grain refinement for improved joint strength and ductility, as well as the avoidance of defects.
The influence of standing ultrasonic waves in laser welding – e.g. by performing welds at different points along the standing wave – is therefore currently being investigated at the Laser Zentrum Hannover and the Leibniz University Hannover.
Experimental setup and procedure
The welding setup being used (see Figure 1a) consists of: a diode-pumped 1,030nm disk laser (TruDisk16002, Trumpf, Germany); welding optics with an optical fibre diameter of 200µm, a collimation length of 150mm, a focal length of 300mm and a focal spot diameter of 400µm; a robot (KR 60 HA, Kuka, Germany); two flat nozzles to supply argon shielding and process gas; and a hydraulic clamping setup.
Figure 1: a) Experimental setup; Top views and metallographic cross sections of dissimilar welds b) Without ultrasonic excitation and c) With ultrasonic excitation
A bespoke system is used to produce an ultrasonic standing wave with a frequency of approximately 20kHz. Welds can then be performed at varying positions along the ultrasonic standing wave. See [1] for a detailed description of the setup.
As part of our investigation, round bars made from 1.4301 steel alloy and 2.4856 nickel-base alloy, with a length and diameter of 30mm, were butt-welded. Welds were performed with the melt pool located at the displacement- node and antinode of standing waves (see Figure 2) with amplitudes of 0, 2 and 4µm. This was done using 7.75kW laser power and at a welding speed of 0.95m/min. The weld shape, microstructure and defects were evaluated and are shown in Figures 1b and c. HV1 hardness measurements were conducted 6mm below the specimen surface. Furthermore, scanning electron microscopy (SEM)-analyses (Quanta 400 FEG, Thermo Fisher Scientific, USA) with EDX-linescans and mappings were conducted laterally to the line of hardness measurement.
This experiment enabled us to assess the influence of ultrasonic excitation on grain refinement, material intermixing and overall weld strength.
Results
It was found that at the displacement node of the ultrasonic standing wave – where maximal pressure occurs – an equiaxial grain structure forms, around which a higher risk of cracking arises. At 2µm standing wave amplitude, fluctuations of the Fe- and Ni-content in the weld take place corresponding with newly formed grains. At 4µm amplitude, the equiaxed grain structure forms in the centre of the weld, corresponding with a different chemical composition than the weld border (consisting of directed grains).
Figure 2: Welds were performed at the various nodes along an ultrasonic standing wave to assess the influence of ultrasonic excitation
At the displacement-antinode of the ultrasonic standing wave – where minimal pressure occurs – it was found that the edges of the melt pool vibrate in such a way that cracking is reduced in the weld. At 2µm amplitude, the chemical composition of the weld is homogenised, with grains of new orientation being found that minimise the hardness difference between the weld and the base material. At 4µm amplitude, V-shaped composition fluctuations occur due to the stronger vibration of the weld edges, moving the melt out of the weld. The number of grain orientations increases and dendrite fragmentation is fostered, presumably due to acoustic cavitation.
Between the displacement node and antinode of the standing wave, it was found that a cross flow occurs in the melt pool that increases with higher ultrasonic amplitude. Consequently, as the weld width increases, the weld depth decreases, destabilising the keyhole.
Conclusion
It was found that the weld properties between round bars made from 1.4301 steel alloy and 2.4856 nickel-base alloy can be improved by performing the weld at the displacement-antinode of an ultrasonic standing wave with an amplitude of 2µm. The chemical composition of the weld is homogenised by the ultrasonic mixing effect, while the hardness gradient is reduced by ultrasonic grain refinement and solution strengthening. The overall weld hardening effect is reduced at higher standing wave amplitudes by porosity (when at the displacement-node) or by a change of melt flow (at the displacement-antinode).
In future investigations, a fully refined and ductile weld could likely be achieved by modulating or combining different types of ultrasonic excitation.
Funded by the Deutsche Forschungsgemeinschaft (DFG, German Research Foundation) – CRC 1153, subproject A3 – 252662854. The authors would like to thank the DFG for the financial and organisational support of this project.
Jan Grajczak is a research assistant at the LZH.
Christian Nowroth is a research assistant at the IDS.
Sarah Nothdurft is head of the Joining and Cutting of Metals Group at the LZH.
This articles was co-authored by Jens Twiefel and Jörg Wallaschek, of Leibniz University Hannover, and Jörg Hermsdorf and Stefan Kaierle, of Laser Zentrum Hannover