Investigation of cavity modes using sound analysis

When high speed oxygen jet hit the molten bath, a cavity is formed. Depending upon the trajectory of splashing, three cavity modes have been defined as dimpling, splashing and penetrating. Though there are various studies that showed the existence of splashing and penetrating modes and how it relates to the droplets amount, no study can be found in literature that focused on the identification of cavity modes using sound approach. In the present work, a novel approach of the sound analysis was carried out on the produced sound in the bath. A 1/10th air-water model was used for this purpose. A linear Pulse Code Modulation (PCM) recorder was kept fixed at certain distance from the bath. Recordings were carried out for 10 seconds in various operating conditions which includes dimpling, splashing and penetrating cavity modes. In order to understand the effect of slag on the produced sound in steelmaking condition, sound analysis was also carried out when a second layer made of motor oil was put over the water surface. Recorded sound was analysed in a computer using open source software called 'WaveSurfer' and 'Audacity'. In 'Wavesurfer', waveform analysis of dimpling, splashing and penetrating modes was carried out. Analysis showed that amplitudes of sound increased as the cavity mode changed from dimpling to splashing and splashing to penetrating (as shown in Figure 1). Using a second layer did not change this finding. In 'Audacity', recorded sound was investigated by spectrum analysis. Interestingly, it was found that at a certain range of frequencies, there was much difference in the sound level as the cavity mode changed from splashing to penetrating (as shown in Figure 2). Existence of a second layer only altered the frequency at which the maximum difference in sound level was found. The findings from the experiments with second layer were consistent with the results when no second layer was used. Though this novel approach showed promising results, it needs work in order to use it in industrial application. It can be predicted from the present analysis that sound approach might be useful in predicting cavity modes