For Case 4.B the radial analysis showed the existence of a large off centered low velocity vortex region at 0.40 D with the presence of at least two separate weak, but distinct, vortex structures. Both vortices merge as a new stronger one after 0.5 D, where the PVC stops spiralling. An observed difference to the unconfined case is that the strength of the new vortex did not decay but increased as it approached the final pyramid exhaust. The PVC passes through CRZ2 and then merges with other structures with a main central wobbling vortex core. CRZ1 is located in the exit of the swirl burner and clearly extends well upstream into the burner nozzle.

 Fig. 16. Different phase angle sections with CRZ1 and CRZ2 in Case 4.A. (A) CRZ1 and CRZ2, 0.00; (B) CRZ1 and CRZ2 attached, 146.25; (C) CRZ2 only, 22.50. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

In Case 4.C, the existence of an extremely powerful vortex was observed in the ‘‘radial” analysis. The PVC always starts from the burner exit with a spiralling movement, stopping its precession suddenly at 0.40 D. No second vortex was observed inside of the PVC. Maybe the second vortex of the previous experiment was only a manifestation of the decrease of the precessional movement and realignment of the vortex towards the central axis. The

 Fig. 17. Axial view of two confined experiments. (A) Strong CRZ apparently split into a strong and weak structure, Case 4.B, 0.00. (B) CRZ1 is anchored in the swirl burner exit, Case 4.C, 45.00. Colors defined in Fig. 16. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

mechanism appears to be similar to Case 4.B. The interaction of two vortices generates a new strong vortex structure which only precesses at lower sections and stops precessing further downstream. An external or corner recirculation zone (ERZ) was found as in all other confined cases and has been ignored. Strouhal number analysis of the PVC frequency showed the following results at high Re.

Case Asymptotic Strouhal number

UNC_1(unconfined) 0.83

4.A and 4.B 0.75

4.C 0.78

Maximum variation is 10% indicating how the mechanism of precession is influenced by the confinement into which the swirl burner is fired.

3.4. Confined conditions. Circular case

The ‘‘axial” plane was analysed first for the open confinement, 4.D. Some distortions appeared in the system compared to earlier results. Both CRZ1 and CRZ2 were found, although they were more closely merged than in the previous cases, for similar swirl burner/ furnace arrangements [13,25,31].

CRZ1 detaches from the burner exit, moving upwards in a more pronounced manner than with the square confinement, 4.A. In the first section both CRZ1 and CRZ2 could be identified. However, between phase angles 90 and 135 a process of eddy unification seems to take place. CRZ1 is lifted to a point where it joins CRZ2. Moreover, the region of reversed flow has considerably increased in size compared to that in the square confinement. Fig. 18 shows the structures at three different phase angles and the three-dimentional nature of the CRZ is evident. The position of CRZ1 is antiphased between the first and last picture because of the rotation required for the analysis. The CRZs are much larger, doubtless due to the circular nature of the confinement.

The ‘‘radial” analysis revealed the precessing vortex core. Although the PVC appears as a very strong structure with clear coherence in the first section, it virtually disappears after 0.41D, suggesting that the vortex has entered into the region of CRZ2 whose strength overcomes the vortical movement of the vertical structure.