Furthermore, the velocity distribution above the first three outlets
demonstrates the biggest conformity to the velocity profile
received from the research. In the fourth outlet there is discrepancy
between profile shapes. However, measured velocities are
diversified depending on its location in relation to the air duct axis.
In the last outlet the real velocity is close to zero and steady for
the whole cross-section because of the fact that there is 0.5m off
the enclosed end of the air duct. What is more, the average velocity
decreases in conjunction with the increasing distance from the
plenum box which is also connected with decreasing flux of air in
the duct (cf. Fig. 9). Mean velocity values decrease in every next outlet along the duct axis due to mass loss. Velocities in points next
to the outlet are significantly larger (1.54–6.16 times) in relation
to average value from the remaining 6 measuring points, which is
caused by the outflow through the outlet. Due to this fact the radial
component of velocity (perpendicular to the duct axis) increases.
This is the largest difference between results of measurement and
mathematical models because the calculations do not consider the
outlets.