African Fusion June 2019

Experimental procedures The objective of the study was to repair the Inconel cladding as follows: 1. Machine the original Inconel clad surface to remove dam- aged material to a minimum remaining clad thickness of 1.5 mm. 2. Develop a cost-effective repair welding solution in terms of welding time and consumable efficiency. 3. Demonstrate that the welding procedure would provide sufficient fusion. 4. Limit the heat input so that the peak temperature in the carbon steel substrate did not exceed the A c1 temperature of the underlying carbon steel, which would ultimately reside less than 1.5 mm from the fusion interface of the repair – Figure 1. The choice in welding process was based on several factors. Of the factors considered, productivity, process control, and level of automation were deemed to be the most im- portant factors that affected the selection of the process. Ultimately, pulsed gas metal arc welding (GMAW-P) was the preferred welding process over gas tungsten arc welding (GTAW). The consumable used to perform the repair was ERNiCrMo-3, which is the welding consumable equivalent grade of Inconel 625. The shielding gas used in the repair was a proprietary mixture. Figure 1: A schematic representation of a cold repair showing the two interfaces and the respective maximum temperature fields allowed at these interfaces. Estimation of weld thermal cycle in HAZ The Rosenthal heat-flow equation is widely used to approxi- mate temperature fields that arise from a weld thermal cycle. Several important assumptionsweremadeduring the formula- tion of the equation, and include: • The heat source is a point source of infinite temperature. • Latent heat due to phase changes is disregarded. • Thermal properties are unaffected by a change in tem- perature. Given these assumptions, the Rosenthal equation can only be used with some confidence to predict the weld thermal cycle in the heat affected zone. In the current study, the Rosenthal equationwas used to estimate thedistancebetween the fusion line (peak temperature equal to the liquidus temperature of Inconel 625) and the original fusion line between the original clad layer and the medium carbon steel, where the peak temperature should not exceed the austenitic transition tem- perature, A c1 , so that austenitisation of the carbon steel does not occur during theweld thermal cycle. It is important to note that the Rosenthal equation cannot be used to determine the penetration depth of the weld into the substrate. A preheat temperature (T 0 ) of 70 °C was assumed as this would be the maximum interpass temperature. The Rosenthal equation, relating a specified peak tempera-

ture to the distance from the heat source, assuming a thick plate, can be seen in Equation 1.

e 2 2 π ρ / q v r  

− = 

(Eqn.1)

T T p o



= peak temperature (K) = preheat temperature (K)

T p T 0

q/v = heat input (J.m-1) ρ

= volume thermal capacity (J.m -3 .K -1 ) r = radial distance from heat source (m) The maximum temperature in the medium carbon steel (T p ) was assumed to be 735 °C, which is at the lower end of the quoted range of 735 to 747 °C [1]. This represents a conser- vative estimate as some overheating is necessary before austenitisation occurs. Because the thermal properties of Inconel and carbon steel differ, the results of the calculations will be sensitive to the value assumed for the volume thermal capacity. It is likely that the true thermal properties of the heat affected zone dur- ing repair welding fall between those stated for Inconel and medium carbon steel. In order to simplify the analytical solu- tion, the Rosenthal equationwas evaluated using the thermal properties of carbon steel and Inconel 625 separately, as shown in Table 5 [1]. It was found that Inconel 625 thermal properties posed a slightly higher risk for exceeding the A c1 temperatures at variedheat inputs (altered through changingwelding speed) as can be seen in Figure 2. The heat input was estimated using an average welding current of 125 A, an arc voltage of 25.5 volt, and an arc ef- ficiency η of 0.8. The welding speed was varied between 200 and 600 mm/min, for purposes of calculation.

Material

Carbon steel

Inconel 625

Volume thermal capacity ( ρ ) (J.m-3.K-1)

4.5 x 106

3.46 x 106

Table 5: Volume thermal capacity of carbon steel and Inconel 625 used in the Rosenthal equation.

Figure 2: Estimated distance between the fusion line (peak temperature 1 350 °C) and the position with a low risk of re-austenitisation (peak temperature 735 °C) using carbon steel and Inconel 625 thermal data from Table 5.

13

June 2019

AFRICAN FUSION