OCR Text |
Show the duct cross section. Passing through each converging wall are two rows of tubes through which air is injected at an angle of 120° relative to the incoming gas flow. Each wall has ten 78.0mm and nine 40.9mm diameter air tubes which can be individually capped to provide many different injector arrangements. The secondary combustor model, shown in Figure 2, was a 203.2mm x 203.2mm square duct constructed from 6.4mm aluminum plate with 3.2mm thick glass windows on one side to allow flow visualization and laser velocimeter measurements. The model was built with separate inlet, mixing and outlet sections so the degree of area restriction in the mixing section could be varied. Three mixing sections were tested, with area reductions of 50%, 25% and 0% (no restriction). Seven holes were located on each converging wall to inject water at either 90° or 120° relative to the bulk flow direction. The hole diameters were adjusted to give the desired velocity ratio between the primary (p) and secondary (s) streams. The five injector/cross section arrangements that were tested are shown in Figure 3. Based on nominal LMF test conditions, a typical mass-flow-rate ratio (ms/mp) for the LMF secondary combustor was found to be fairly constant at about 0.7 5, so the mass ratio in the model was fixed at a constant value of 0.7 5 in all tests. Typical bulk velocity ratios (Vs/Vp) for the LMF secondary combustor were found to range from about 10 to 20, so the model tests were conducted with velocity ratios of 10 and 20. In the model study, the primary flow rate was set at 0.0038 m3/s, resulting in an inlet bulk velocity of 0.092 m/s and an inlet Reynolds number of 1.65 x 104, which is about 30% of the value in the real secondary combustor. The secondary flow injection rate was set at 0.0028 m3/s. Test conditions for the five cases studied are summarized in Table I. INSTRUMENTATION The flow field was measured with a two-component Bragg diffracted laser velocimeter (LV) system, which was designed and constructed at UTSl(3). The LV measures velocities in a flow by detecting the frequency of intensity variations in light scattered by particles as they pass through the interference fringes produced by two intersecting coherent laser beams. The two beams are split from one laser beam using Bragg diffraction, which induces a frequency difference fb between the intersecting beams. This causes the fringes to move so that a stationary particle in the beam 5 |