Wire Arc Additive Manufacturing (WAAM) uses an arc welding process and consumable filler wire to deposit near net shaped 3d structures in a layer by layer manner. The highly flexible nature of the process enables the production of components with materials that are otherwise challenging to process using subtractive manufacturing methods. Invar 36 is a low thermal expansion Fe-Ni alloy deriving its name from the invariability of the thermal expansion to temperature, which is the lowest at a nickel content of 36 wt.%. Invar is challenging to machine due to a low thermal conductivity, high toughness and high susceptibility to work hardening, resulting in low productivity and high machining costs. Consequently, additive manufacturing methods such as WAAM can provide a good alternative to conventional production processes.

The processing parameters required for defect-free deposition of a commercial invar 36 welding wire using GTAW-based WAAM were investigated in this research. Multi-layer structures with a target cross sectional area of 15x15 mm2 were deposited using processing parameters corresponding to heat inputs ranging from 200- 550 J/mm. Microstructural evaluation revealed a coarse columnar grain structure with a mostly cellular mixed mode solidification substructure. Additionally, an increasing heat input resulted in a higher concentration of voids and (micro)cracking along the grain boundaries. Ductility-dip cracking was the most likely cracking mechanism due to the observed locations and fracture surface of the cracks. Defect-free invar weldments were deposited at a heat input of 200 J/mm.

Two 20x120x60 mm3 invar 36 deposits were made at 200 J/mm and 550 J/mm to allow for the machining of tensile testing samples. A higher average strength and elongation was observed in the 200 J/mm heat input samples when compared to the 550 J/mm samples. Samples oriented horizontally in the block displayed a 5.2 % higher UTS and a 8.5 % lower Young's modulus. The highly directional solidification grain structure resulted in anisotropic behaviour due to the Hall-Petch relationship and elastic anisotropy of the austenite crystal. A high concentration of grain boundary defects caused premature failure in 75 % of the 550 J/mm samples.

Multi-wire WAAM, where both invar 36 and nickel welding wires were used, was used to deposit invar alloys with a variable nickel content. A stepwise graded wall was deposited, where the invar composition was increased every three layers ranging from 33-38-42-48-55 wt.% Ni. Energy Dispersive X-ray Spectroscopy (EDS) analysis showed a homogeneous composition in each layer and a gradual increase of the nickel content throughout the wall. A defect-free invar 42 block was deposited at a heat input of 215 J/mm. The nickel composition was found to be homogenous throughout the sample and in accordance with the model created to calculate the weld metal composition as a function of the individual wire feed speeds. X-ray diffraction (XRD) analysis of all invar compositions deposited in this research showed a diffraction pattern matching the fcc diffraction pattern.