Rd. Flack et al., MEASUREMENT AND PREDICTION OF NATURAL-CONVECTION VELOCITIES IN TRIANGULAR ENCLOSURES, International journal of heat and fluid flow, 16(2), 1995, pp. 106-113
Free convection velocities were predicted with a finite element method
and measured using laser velocimetry in three isosceles triangular en
closures with base angles of 30 degrees, 45 degrees, and 60 degrees. T
hree Grashof numbers were tested for each geometry, which ranged from
1.89 x 10(6) to 10.3 x 10(6). Velocity data were measured near the two
isothermal side walls and the isothermal bottom wall. Data were nondi
mensionalized by the same parameters used to nondimensionalize incline
d plate data. Flows were laminar for ail cases. For a given base angle
, the nondimensional data and predictions collapsed to single curves b
est for either of the side walls with typical variations of 5 percent;
nondimensional predictions and data collapsed poorly for the bottom w
alls, with typical variations of 20 percent. Hot and cold side-wall da
ta exhibited similitude and typical differences between nondimensional
results for the two side walls were 10 percent. For varying base angl
es, side-wall results failed to reduce to single curves with variation
s of typically 50 percent. Absolute differences between predicted and
measured peak velocities ranged from 6 to 35 percent. Differences betw
een predicted and measured velocities for base angles of 30 and 60 deg
rees tended to be positive, whereas, differences for a base angle of 4
5 degrees, tended to be negative.