Dear Oliveria thanks for your valuable answer, but I have small doubt can we really use GTAW for additive manufacturing since this process does not include consumable welding wire.
GMAW certainly can be used - this is sometimes known as Wire+Arc Additive Manufacturing (WAAM). A lot of work on this has been done recently at Cranfield University (among others).
GTAW can be used for AM either with a wire feed attachment or with a blown powder feed. As you mentioned, with autogenous GTAW there is no source of feed material, so that would not work.
Thanks kornel for your answer actually you are right I am working on that thing only, I have fanuc robotized GMAW process and I want to use this process as a metal 3 D printer.
Much more information on waammat.com which is the outlet of the research Harry Coules was referring to :)
The main issue is not the process itself, but how to choose the best parameters and tool path to avoid defects such as lack-of-fusion, and keeping the layer height under control.
It takes a few days to find the parameters to build a simple linear structure based on single beads, but months if not years to learn how to make a proper engineering component with a suitable level of structural integrity
Thanks Dear Martina for your answer, I have followed Harry Coules reference and I found it really interesting. Now in initial phase I am trying to get the linear structure with single beads.However, for multi-pass beads the control of arc force at the desired location is quite difficult.
For multi-pass beads the possibility to control the heat input and the wire feed speed independently from each other is absolutely key, and in GMAW you can't do this I am afraid. Indeed for more complex geometries we prefer to use plasma
Thanks Dear Martina for your response. Yes, you are right, the main thing is to control the heat input and the wire feed rate, independently. I have successfully controlled both these parameters independently in SAW process, now I am working on GMAW. I think that will help in multi-pass beads.
I think, the use of an auxiliary power source for the preheating of welding wire separately will help in controlling the heat input at the desired rate while depositing the multi-layers.
You need almost on-the-fly change of parameters with immediate response, you need geometry-based compensation every time you come to intersections or changes of features, remember, in DED keeping the layer height constant (and what you want it to be) is absolutely fundamental and is the most difficult part
I don't think the approach you're suggesting has got enough response timeliness and it's rather long-winded when you could just switch to plasma or TIG and play with the current in real-time