| OCR Text |
Show 5.6.10 adjusted from the tangential position to an angle of 60° inward from tangential. It was found that for all cases of inlet number and angle, the flow within the chamber was substantially the same. As the number of inlets were reduced in number the flow became less axi-symetric in the outer regions of the chamber radius, however, no marked change was observed in the main body of the chamber. In the limiting case of only one inlet, the disturbance was only apparent in the outer 5% of the chamber radius. When operating on two or more inlets, the effect of altering their angle, to point more towards the centre of the chamber, was examined. Once again the change in flow patterns was limited to the outer regions of the chamber, with the flow streamlines soon bending round to follow a near tangential path. It was only when the inlet angle exceeded 60%, that the flow pattern changed substantially, by changing to a small vortex flow in the centre region of the chamber, with a large area of stagnant flow around the outside. In appearance it was similar to the flow patterns in a 'corner-fired' power station boiler. The large region of stagnant flow was considered to be a waste of space and this configuration was not investigated further. The vortex flow pattern proved to be an extremely stable phenomena and even major changes in the driving arrangement of the vortex combustion chamber had little effect on the resulting flow patterns. It has already been mentioned that the effect of chamber size had no effect on the form of the profiles obtained, therefore the next part of the investigation was aimed at finding the effect of changes in the form of the chamber, expressed as a ratio of primary chamber dimensions. The first series of tests were performed on a range of chambers with different length to diameter (L/D) ratios. Again very little change in the overall pattern of the flow was observed. As the chambers became longer and thinner, L/D increasing, the tangential velocity profile tends to peak slightly more towards the axis of the chamber. This can be explained by the proportionally larger amount of air flow which has to be squeezed into the central axial region before it can leave the chamber, as the L/D ratio increases. A ten-fold change in L/D from 0.2 to 2.0 changed the flow profiles by less than 10%. The most striking change in flow profiles occurred when the exit diameter of th<? c:hamhor (Do) wan changed. This was expressed as a ratio with chamber diameter (De/D). It was very obvious that this determined the radial position in the chamber where the peak tangential velocity occurred (Figure 4). This peak velocity occurs at the change-over from a fixed vortex core, to the outer free vortex flow, and has been observed to be the position of high turbulence and good combustion8. In our experiments it was found that a number of |
