Metal vaporization from weld pools
- PDF / 648,919 Bytes
- 9 Pages / 603.28 x 783.28 pts Page_size
- 10 Downloads / 250 Views
I.
INTRODUCTION
Loss of alloying elements from the weld pool due to vaporization is important for a number of reasons. Firstly, if the loss is great enough, the mechanical properties of the weld may be impaired.L Secondly, the composition of a welding arc plasma influences the temperature of the arc, 2 arc stability, and fume formation. 3 Thirdly, it has been shown that vaporization places an upper limit on the temperature produced on the surface of the metal due to evaporative cooling. 4,5 A previous paper has presented a formalism for calculation of partial pressures of metal vapors above steel weld pools. 5 This analysis provided an estimate of the power lost by evaporation as well as an upper bound on the surface temperature of steel weld pools as a function of alloy composition. In the present paper, this analysis is extended to evaporation from aluminum and copper alloy weld pools where different metal vapors dominate. The results of the calculations are then compared with experimental results from both aluminum and steel weld metals. II.
ARGON ---~
.
i
Tt
I / SwPeCtromet~er Welectrode
i
j
EXPERIMENTAL PROCEDURE
Both steel samples and aluminum alloy samples were subjected to extended time welding in a specially adapted chamber which was coupled to a direct reading emission spectrometer. The chamber was supplied with a lens guiding the light to the spectrometer grating. A mirror imaging device was provided to ensure uniform position of the work and the electrode. The hearth was water cooled and the entire system was purged with argon flow. The rotating water-cooled copper hearth shown in Figure 1 was employed in the case of 12.5 cm diameter 304 steel samples, but the rotation has not proven useful. The slight deviations of the rotating sample cause periodic sinusoidal changes in the arc length and corresponding oscillations in the spectral signal. Therefore, this experiment could be characterized only by lower and upper limits of spectrographic signal, voltage, current, and final weld chemical composition, and not by a uniquely defined value of each of those parameters. As a result, the experimental data presented in this paper A. BLOCK-BOLTEN, Research Scientist, is now with the Department of Metallurgy and Materials Science, University of Toronto. T.W. EAGAR, Associate Professor, Materials Engineering, is with the Department of Materials Science and Engineering, Massachusetts Institute of Technology, Cambridge, MA 02139. Manuscript submitted September 23, 1983. METALLURGICALTRANSACTIONS B
" ~ - - - ' - - ~'
-.
WATER INLET
Fig. l--Water-cooled turntable and gas-cooled electrode holder.
were obtained from stationary arc welds where the arc length was more controllable. The 0.35 to 1.10 g steel samples consisted of 1.6 mm diameter wires (502,505, 5151, and 5212 steels) or 1.1 mm wire (410 steel) or 2.4 mm diameter wires (308L and 309L steels). The thoriated tungsten electrodes of 1.6 mm diameter were mounted as shown in Figure 1. Each steel sample was weighed before and after each experiment, ye
Data Loading...
