Never before has an optical method been used to determine the arc temperature inside a submerged arc welding cavern with frequencies up to 5 kHz. To be able to do that, a combination of high-speed imaging and spatially resolved high-speed spectroscopy with up to 5000 fps has been performed. A DCEP (direct current electrode positive) process with 600, 800, 900 and 1000 A, and an AC (alternating current) process at 800 A were included in this research. The Bartels method has been used to calculate these temperatures for the first time. It generated temperatures from approximately 7000–9000 K. Also, it was found that a decrease of arc temperature for rising currents appeared until 900 A. After that, it is reversed for currents higher than 900 A.

Recent research of gas metal arc welding (GMAW) has proved that the sheath voltage dominates the total voltage fall in the current circuit and provides the most energy to the wire and the weld pool. In this paper, the energy delivered to the wire anode is studied experimentally for a typical pulsed GMAW process in the one drop per pulse mode to provide a more complete picture of the energy balance of the process. Welding of mild steel under Ar with 2.5% CO ₂ with different peak currents from 350 to 650 A and constant wire feed speed of 4 m min ⁻¹ as well as constant arc height are considered. The processes were studied with non-intrusive optical methods, i.e. without strong modifications. The droplet energy was determined by high-speed two-colour pyrometry and an analysis of the droplet surface temperature and geometry. In addition, Joule heating in the wire anode was also estimated from surface temperature measurements. It was found that the averaged droplet energy is almost independent of the peak current and approximately one-third of the total electric energy. A voltage equivalent of the energy delivered to the wire anode from the plasma was defined. Substracting the work function of iron, the value of this voltage is in the range of 3.25± 0.43 V and increases by 0.5 V when the current increases from 350 to 650 A. The sum of the cathode and anode sheath voltages have been estimated to be in the range between 18 and 23 V in preceding works.

A first experimental validation of the evaporation-determined arc-cathode coupling (EDACC) model is performed by comparing the experimental and simulated current in the peak current phase of a pulsed gas metal arc welding process. For this, the EDACC model was extended to limit the cathode surface temperature to a realistic value of <2400 K. The information on the plasma for the EDACC model was gathered from literature and extrapolated and extended according to qualitative reasoning. The information on the cathode surface of the EDACC model was derived from a steady-state simulation of the weld pool, using an averaging approach over time for the energy and current. The weld pool surface temperature was compared to pyrometric measurements, that were performed for this work, and the agreement was found to be fair. The observed agreement between the modeled and experimentally determined current was within 10%. As strong assumptions were made for the comparison, the validation cannot be considered as final, but the assumptions are thoroughly analyzed and discussed. However the critical link between surface temperature, plasma temperature and total current transmitted could be reconstructed.