Influence of Hydrogen Shielding Gas Additions on Base Metal Dilution, Bead Shape and Wall Buildability in Fe-10Ni Gas Metal Arc DED
Michael Moore and Dennis D. Harwig
Ohio State University
A new high strength steel designated Fe-10Ni is being developed for potential use in as-welded Naval ship structures. Prior art has shown the weld properties are dependent on oxide particle morphology and grain size and in multi-pass gas metal arc welding pulse (GMAW-P) deposits. Prior research evaluated the effects of Argon-O2 and Argon-CO2 mixtures effects on as-welded strength and toughness. The oxidizing shielding gas additions were necessary to provide good wetting and metal transfer weldability. For Fe-10Ni welds, best properties had the lowest concentration of oxide particles. Hydrogen shielding additions provide a reducing environment, improve pool fluidity and may offer “high duty” directed energy deposition (DED) of structure by minimizing scale build-up associated with oxidizing shielding gases. Using a level believed to minimize hydrogen cracking risks, the effect of 98 Argon - 2% hydrogen (H2) shielding gas on Fe-10Ni GMA DED deposits were evaluated for “single-pass per layer” walls and “multi-pass per layer” blocks. Bead shape and base metal dilution relationships were determined and compared to deposits made using Ar-5%CO2 to produce sound walls and blocks and evaluate properties. Wall and block strength and toughness were characterized using tensile and Charpy impact tests for two different deposit sizes.
Fe-10Ni (1.2-mm) welding electrode was used to develop wall and block GMA-P DED procedures using both Ar-5%CO2 and Ar-2%H2 shielding gases at 35CFH flow rate. A Cloos robotic welding system was used to develop procedures for two deposit sizes for each build condition. Two deposit sizes were fixed using a wire feed speed (WFS) to travel speed (TS) ratio of 15 and 30. For each WFS/TS ratio, a series of wall and block tests were performed to evaluate the effects of power (controlled by WFS level) on bead shape and base metal dilution and determine an operational window for acceptable build parameters. Build tests were performed on 1-inch thick by 3-inch wide A572 Grade 50 plate build platform. The builds were oriented on the 1-inch width as a distortion control mechanism. The first six layers were used to eliminate platform dilution effects, establish a steady-state 200C interpass temperature and create a steady-state “heat sink” condition for build condition evaluation. A series of constant deposit area tests were performed using WFS levels from XX to XX. The contact tip to work and arc length were kept constant based in operator preference for pool control for each deposit size. The wall and block test weld were evaluated in the as-deposited conditions and then section to measure bead shape characteristics. The as-deposited evaluation included a photograph of surface scale and a measurement of minimum and maximum bead width due to any undulations along the wall and block corner bead deposit length. The bead width, bead throat and base metal dilution of the final deposit were measured for each wall and block build test. Graphs were produced and used to evaluate the effects of shielding gas and power on as-deposited and cross-section bead shape quality.
Results and Discussion
Finished builds were removed from the build platform as well as 1-1/2-inches from the start and end of each build. Locations for tensile, Charpy, LOM, LECO and Auger tests were designated in Figure 1.
Tensile and Charpy impact samples were produced using EDM and the results are shown in Figures 2, 3 and Table 1 respectively.
Tensile results were comparable to HY 130 plate. Additional testing is being conducted on 4-deposit per layer blocks of Ar-5% CO and 4-deposit per layer wall/block of Ar-2% H2 samples. Each gas and wall/block configuration will be tested utilizing two different deposit sizes.
Elongation and area reduction results from the first set of tensile test samples demonstrate excellent ductility, which is promising for defense applications.
Charpy impact numbers in Table 1 demonstrate a lack of toughness relative to Fe-10Ni plate. Hydrogen additions to the shielding gas is expected to reduce the #density of oxide inclusions. Further parameter development is in process to reduce grain size. The combination of these two factors has been shown to significantly improve the toughness of deposited Fe-10Ni material.
Block builds made consisting of 4-deposits per layer are shown in Figure 4 and demonstrate the edge retention quality and undulations present in the initial block build with Ar-5%CO2.
Silicone oxide buildup as compared to nominal oxide presence is shown in Figure 5. Furthermore, the bead shape contrast is evident between the two shielding gases. A 2% hydrogen addition to argon shielding gas produces a flat bead surface in contrast to the more convex bead profile with a 5% CO2 addition.
Ongoing experimental work is being done to formulate the scientific mechanisms behind hydrogen additions that influence the bead shape (width, throat) and base metal dilution from previously deposited layers. This work will also test the theory that oxygen additions to shielding gas is the leading contributor to the # density and size of oxide inclusions. Developing consistent builds for microstructure characterization require adaptable GMA-P DED-AM parameters that maximize edge retention, base metal dilution control and minimize the number of deposits due to the added complexity of the thermal cycles inherent AM. Specifically, the influence of hydrogen additions on the melting rate, droplet growth behavior and droplet transfer are being formulated for Fe-10Ni filler material. A high-speed camera/DAQ setup focused separately on the electrode and the molten pool are in process. Completed results by June 2023.