OCR Text |
Show The majority of the velocity measurements reported herein have been acquired with a singlechannel LDA. The system is constructed from components supplied from three major sources: the 3W Argon Laser is from Spectra Physics; the optics and Bragg cell frequency shift unit is from Disa; and a Harwell filter bank and software package was used for data processing. The system operates in the back-scatter mode and is operated on a 3-D traversing table. COMPARISON OF MODEL AND EXPERIMENT EXISTENCE AND POSITION OF THE IRZ - The ability of the inviscid theory to predict the critical swirl number required for the formation of an internal recirculation zone and a comparison of the axial location of the IRZ between theory and experiment is shown in Figure 8. The data demonstrates the relationship between swirl level and the extent the vortex must expand before an IRZ is formed. The diameter expansion ratio refers, in this case, to the local radius of the quarl relative to the throat radius at any axial position. As can be seen in Figure 8, an IRZ was found experimentally down to a swirl number of 0.4, while the inviscid theory predicts that the critical number would be 0.5. However, after the IRZ is formed, the axial position (or expansion diameter ratio) agrees very well with inviscid theory. o ;: c a: z o W z C Go )( W a: w ~ w ~ ~ o PREDICTED CRITICAL SWIRL NO. 20~----~ ___ 1r---______ _ 1.5 +------.:o+--+-----O-Ol-. -20-:-'~ NO IRZ FORMED MEASURED CRITICAL SWIRL NO. a I i~ I INVISCID PREDICTION a ~ .35' fr.e ,et B/A .15 and B/A .20 1.0+-----..L.L.....a--+------_~ o as 1.0 SWIRL NUMBER S Fig. 8 - Comparison of inviscid theory prediction and experiment for the axial position of the IRZ versus S VELOCITIES IN THE QUARL - Figure 9 shows a comparison between the measured U and W velocity profiles when the flow expands in a quarl and those predicted with inviscid theory, equations (4) and (5). As can be seen for this case, the agreement is quite good, and the inviscid boundary condition [1] for expanding, swirling, simple flows (no bluff body) of U = W = 0 everywhere in 94 U/U. W/ Ue tS t--~f-----+----I LOr~1I) l 1)(: 200 ] I 1 _ tneorP1.c.aI ~~g l.,qlt'f·..-1Ij .. .. ~.-- -- ....... .. I .1 1':0 1\ I r II ltV ~ . : Fig. 9 - Velocity profiles predicted by the inviscid equation SBR, S=0.7, B/A=2 ( <F20C), Df = ~40 the IRZ and at the IRZ boundary is also experimentally verified. EFFECT OF THE INITIAL SWIRL LEVEL AND THE DIAMETER EXPANSION RATIO B/ A ON THE FLOW IN THE IRZ - Inviscid theory implies that flow fields should be similar when the initial swirl number and the amount by which the fluid is expanded is held constant, independent of the manner in which the expansion occurs. Figure 10 shows the IRZ boundary and internally recirculated mass flows for a number of quarl geometries, but the same furnace. For these highly confined cases, the recirculating mass flows in the IRZ were almost identical when compared on a diameter expansion basis. In Figure 10, the recirculated mass flows are shown as a function of axial distance and, consequently, the axial position where the IRZ strength reaches a maximum depends on the quarl used. N , " 010 10 H--+--J-......;...-+~ +---+----1 '0 Fig. 10 - Effect of quarl angle and length for high confined swirling flows: SSBR When the flow exits the quarl into a large furnace, the flow in the IRZ is still determined by the distance the flow expands inviscidly, but this distance is no longer fixed by the furnace diameter. As shown in Figure 11, the recirculation zone boundary expands for only a short distance after leaving the quarl into a large furnace and then begins to close. Defining the region of inviscid expansion as the position where the IRZ diameter is greatest, flow similar- |