AN EXPERIMENTAL STUDY OF A PARTICULAR ARRAY OF RECTANGULAR JETS WITHIN RECTANGULAR DUCT

The mixing of a linear array of twenty rectangular jets confined in a rectangular duct has been investigated to determine the effect of the duct height on the flow behaviour. This type of configuration is of importance in the thrust-augmenting wing concept used in V/STOL aircraft, which requires optimum mixing in a minimum duct length.;The primary flow issued from the 4mm x 38mm rectangular nozzles at a nominal velocity of 60m/s, to give a jet Reynolds' number of 14,000. Four duct configurations were investigated, with the duct height set at 1.5, 2.0, 2.5, and 3.0 nozzle heights. This gives a Reynolds' number of order 10('5) based on the bulk velocity and duct dimensions. For each configuration, the following data were obtained: the wall static pressure distribution along the length of the duct; the centre-line behaviour of the mean axial velocity and the axial turbulence intensity; and profiles of the mean axial velocity and the axial turbulence intensity along the transverse centre-plane and the central portion of the lateral centre-plane of the duct at selected downstream stations. For the configurations at minimum and maximum duct height, contour plots of the mean axial velocity and the axial turbulence intensity at selected downstream stations are also presented.;For each configuration, the flow is characterised by three distinct regions in the centre-line behaviour of the mean axial velocity: an initial constant-velocity core, a "characteristic" decay region, and (for the minimum duct spacing) a developing duct flow region. The individual jets exhibit "saddle-back" behaviour along their major axis direction: an asymmetry is observed in this behaviour about the lateral centre-plane, and it is proposed that this is the result of asymmetries in the flow upstream of the inlet plane.;Two distinct flow regimes are observed: one in which mixing occurs well upstream of the duct exit and separation of the secondary flow from the duct walls is possible; the other with mixing occurring for the entire length of the duct, and with no evidence of secondary flow separation.