P 875 – Thick-wire gas metal arc welding of unalloyed steels
Within the project, fundamental investigations on gas metal arc welding (GMAW) with thick electrodes of up to 4 mm have been made. Solid wire electrodes on tubular specimens of low alloy steel with CV- and CCcharacteristic lines have been used to obtain welding beads by welding tests.
For the welding experiments, current and voltage have been recorded and synchronized high-speed imaging was performed. These values have been used to analyse droplet formation, and detachment, as well as arc formation.
Analysis of high-speed imaging showed, that a welding -power related classification, as known from SMAW with wire diameters up to 1.6 mm, in short- ,transition-, and spray arc is not possible with wire diameters of > 3 mm.
A very short arc, placed below the metal sheet surface, has been proved to be beneficial in all power-ranges. Spattering could be reduced by this measure. The weld metal is driven out of the joining zone by long arcs, which emerge from the sheet metal surface. This gives rise to the cutting effect.
Furthermore, it could be observed, that both power source characteristics can be used in general. However, when compared with the CV-characteristic, the CCmode yields a significantly more stable process andsplattering is reduced. Every detachment of a droplet alters the arc length. These changes were compensated by regulations of the power source. The CCcharacteristic regulates the voltage in a way that allows droplet detachment with reduced splattering.
On the other hand the CV-characteristic causes current fluctuations of up to 400 A when the arc length is changed by droplet detachment. These fluctuations can literally blast away droplets leading to increased splattering. Therefore the CC-characteristic has been used for welding the experiments that have been carried out.
Admixing of 30 % Argon into the CO2 caused a further reduction of spattering. Higher Argon concentrations have not proved to be effective. The plasma is dominated by metal vapour as the short arc is placed below the metal surface. It determines the shape of the weld penetration and the droplet detachment. The weld penetration can be influenced by the variation of the energy input per unit length. Assuming a constant welding power, the weld penetration and melting deposition rate can be
increased by reducing the wire diameter. This effect occurs due to the higher current density.
In general it can be assumed that the welding current can amount up to 200 times the wire diameter (e.g. 3.2 mm wire corresponds to 640 A welding current) to maintain a stable process and acceptable weld seams. By that, melt deposition rates of more than 8 kg/h are achievable.
Joint weldings have been carried out with metal plates of 12, 15, and 20 mm thickness. Therefore a G3Si1 electrode of 4.0 mm diameter was used. As no wires of other material qualities have been available, welding tests with S460 M and S690 QL materials have been omitted in agreement with the project supporting committee. Therefore a higher workload was concentrated in process-analysis and -evaluation.
Visual inspection and dye penetrant tests demonstrated, with few exceptions, a high quality level of the ob-tained weld surfaces in respect to undercuts, surface pores and cracks. By radiographic inspection, it was found that pores were formed inside the seam due to discontinuous wire feeding. Nevertheless, pore-free seams could be produced within the project.
To determine the mechanical and technological characteristics, hardness measurements, tensile, and impact tests have been carried out. The results showed excellent properties of the weld metal. The hardness peaks in the heat-affected zone at the fusion line amounted to values of up to 326 HV10. In addition, the impact toughness of the weld metal was found to be far better than that of the base material. In the tensile tests of the welded joints, fracture occurred, with one exception, in the base material. The failure in the joint zone was caused by a pore nest at the seam end, which was created by process instabilities.
With respect to seam formation, it should be noted that, compared to submerged arc welding (SAW), imbrication and shaping of the seam were much less beneficial. In contrast to the SAW process, it is not possible to adjust the seam width by the arc voltage in thick-wire GMAW as the length of the arc has to be adjusted at very short distances. So the seam width is only slightly adjustable by the welding speed. Furthermore, it must be noted that the GMAW-thick wire welding, in contrast to the submerged arc welding, is a very sensitive process in which the smallest process instabilities, such as discontinuous wire feed and stagnation in the movement, cause pores and pore nests.
Continuity in wire feeding and welding movement is therefore essential. At present, a robust thick-wire GMAW technology was not jet achieved. Hence, a revision of the wire feed system used is necessary.
It can be concluded that, when all process specifics were considered and followed, the thick-wire GMAW process can be carried out safely. The investigations have shown that the position / counter-position technique can be recommended. Welds on 20 mm plates could be carried out reliably with welding speeds of 90 cm/min.
The research project (IGF-Nr. 16938 BR) was carried out at Technical University Chemnitz, Institut für Fertigungstechnik/ Schweißtechnik. FOSTA has accompanied the research project work and has organized the project funding from the Federal Ministry of Economics and Technology through the AiF as part of the programme for promoting industrial cooperation research (IGF) in accordance with a resolution of the German parliament.
Only available in german language.
R. Agsten, S. Brumm