Phase-redistribution and full-phase separation data were generated for two-phase (air-water) flow splitting at an equal-sided impacting tee junction with a horizontal inlet and inclined outlets. The flow loop incorporated a tee junction machined in an acrylic block with the three sided having an equal diameter of 13.5 ± 0.1 mm I.D. Both sets of experiments were conducted at a nominal pressure (Ps) of 200 kPa (abs) and near-ambient temperature (Ts). The operating conditions for the phase-redistribution experiments were as follows: inlet superficial liquid velocities (JL1) ranging from 0.01 to 0.18 m/s, inlet qualities (x1) ranging from 0.1 to 0.9, mass split ratios (W3/W1) from 0 to 1.0, and outlet inclination angles ranging from horizontal to vertical. These inlet conditions corresponded to inlet flow regimes of stratified, wavy, and annular. Phase-redistribution data revealed that the redistribution of phases depended on the inlet conditions, the mass split ratio at the junction, and the inclination angle of the outlets. The magnitude of the inclination effect was dependent on the inlet flow regime. The phase redistribution in stratified flow was very sensitive to the outlet angle and full separation could be achieved at angles as low as 0.7°. Wavy flow was less sensitive to the outlet angle and annular flow was even less sensitive to the outlet angle.
The capability of a single impacting tee junction to perform as a full phase separator has been examined. Experimental data were obtained for the limiting inlet conditions under which full separation was attainable at various outlet inclinations (θ) of 2.5°, 7.5°, 15°, 30°, 60°, 75°, and 90°. Full separation data have shown that a single impacting tee junction can perform as a full-phase separator for some inlet conditions. Flow phenomena near the limiting conditions were observed and a simple correlation based on the similarity between these flow phenomena and the phenomenon of liquid entrainment in small upward branches was developed. This correlation was capable of accurate prediction of the data in terms of magnitude and trend.