Naphthalene sublimation experiments have been conducted to study the effects of channel orientation, rotational Coriolis force, and a sharp turn, on the local heat (mass) transfer distributions in a two-pass square channel with a sharp turn and smooth walls, rotating about a perpendicular axis. The test channel was oriented so that the direction of rotation was perpendicular to or at a 45 deg angle to the leading and trailing walls. The Reynolds number was kept at 5,500 and the rotation number ranged up to 0.24. For the radial outward flow in the first straight pass of the diagonally oriented channel, rotation-induced Coriolis force caused large monotonic spanwise variations of the local mass transfer on both the leading and trailing walls, with the largest mass transfer along the outer edges of both walls. Rotation did not lower the spanwise average mass transfer on the leading wall and did not increase that on the trailing wall in the diagonally oriented channel as much as in the normally oriented channel. The combined effect of the channel orientation, rotation, and the sharp turn caused large variations of the local mass transfer distributions on the walls at the sharp turn and immediately downstream of the sharp turn. The velocity fields that were obtained with a finite difference control-volume-based computer program helped explain how rotation and channel orientation affected the local mass transfer distributions in the rotating two-pass channel.

Naphthalene sublimation experiments have been conducted to examine the effects of channel orientation, rotational Coriolis force, ad a sharp turn, on the local heat (mass) transfer distributions in a two-pass square channel with rib-roughened walls, rotating about a perpendicular axis. The test channel was oriented so that the direction of rotation was perpendicular or at a 45 deg angle to the leading and trailing walls. In the two straight passes of the test channel, there were parallel 90 or 60 deg ribs on the leading and trailing walls. The test channel modeled serpentine cooling passages in modern gas turbine blades. The results showed that the heat (mass) transfer was very low on the leading wall of the first pass when the channel was oriented with the rotating direction normal to the leading and trailing walls. There were regions of very low heat (mass) transfer on both the leading and trailing walls in the turn, especially on the trailing wall in the turn when the channel with transverse ribs was oriented diagonally. For the given diagonal channel orientation, rotational Coriolis forces caused the leading and trailing wall heat (mass) transfer to be high near the outer edges of the walls in the channel with transverse ribs; rotation-induced secondary flows dominated near wall rib-induced secondary flows in the channel with angled ribs, since the heat (mass) transfer was generally higher near the outer edges of the walls than near the inner edges in the first and second straight passes. [S0022-1481(00)00201-2]

No abstract

No abstract

Experiments have been conducted to study the effect of rib size on the local heat (mass) transfer distribution for radial outward flow in a rotating channel with transverse ribs on the leading and trailing walls. The test channel modeled internal turbine blade cooling passages. Results were obtained for Reynolds numbers of 5,500 and 10,000, rotation numbers of 0.09 and 0.24, and for a fixed rib pitch that was equal to the channel hydraulic diameter. For a fixed rib configuration on the leading wall, increasing the size of the ribs on the trailing wall increased the heat (mass) transfer on the leading wall. Ribs with D/e = p/e = 16 on the trailing wall performed better than ribs with D/e = p/e = 10. When the rotation number was large, the heat (mass) transfer on the leading wall was quite low, regardless of the sizes of the ribs on the leading and trailing walls. There was very little spanwise variation of the local heat (mass) transfer between the transverse ribs on the trailing wall. When the rotation number was large, however, there was a significant spanwise variation of the local heat (mass) transfer between ribs on the leading wall.