10 Dynamics of Separation Characteristics of Sieving and Flow …
371
could not be proven with LDV since the tracer particles were rarely carried into
this area.
For the radial velocities, the same color code was used. For all three revolution
rates, the radial velocity was negative in the central shaft area as a consequence of the
centrifugal force affecting the tracer particles. Between suction and the inner edge of
the wheel, the airflow is very rectified due to the continuity condition. Between the
blades, the radial velocity varies significantly. Based on a mass balance, the radial
velocity at the outer edge of the wheel is about 7 m s
−1 and increases up to ca.
12 m s
−1 .
While at 3000 rpm a relatively uniform velocity profile is established, which follows the above constructed picture, at 9000 and 15,000 rpm the profiles are much less
homogeneous The formation of the profile is due to the secondary vortex discussed
above.
The deflector wheel considered here exhibits an experimental sharpness of cut
of 0.57 at 3000 rpm, 0.59 at 9000 rpm and 0.50 at 15,000 rpm. Comparing these
observations with the flow fields shown before, the reduction of the sharpness of cut
at high revolution rates is not surprising.
This effect will be included in the calculated separation curves shown in the
following.
A parameter which affects the sharpness of cut is the rotational symmetry of the
airflow. At low revolution rates, radial and tangential velocities at the outer edge
of the wheel are very similar and rather low. It is expected that the airflow is more
symmetrical at higher revolution rates where the circumferential velocity of the wheel
dominates. In Fig. 18 the average tangential velocities are shown on the left and the
radial velocities on the right for different revolution rates. The measurements on the
“North” side are indicated in black (cf. vertical plane at radii of 23 and 26 mm shown
in Fig. 15, right) and the “East” side in red (horizontal plane mostly at a radius of
25 mm in Fig. 15, right).
Generally, it is observed that the tangential velocity is more homogeneous than the
radial velocity which was already found for the North side in Fig. 17. At the different
revolution rates, local minima appear which are not at the same positions for North
and East measurements. This indicates that the vortices in the apparatus affect the
flow around the wheel in different ways. The radial velocity, on the other hand,
exhibits a more pronounced and rather systematic divergence. While on the North
side a mostly inwards oriented flow is encountered, the radial velocity on the East
side is nearly zero up to moderate revolution rates. Only at higher revolution rates,
the radial velocity is slightly oriented outwards. This missing rotational symmetry
needs to be taken into account in the following considerations about the particle
separation.
371
could not be proven with LDV since the tracer particles were rarely carried into
this area.
For the radial velocities, the same color code was used. For all three revolution
rates, the radial velocity was negative in the central shaft area as a consequence of the
centrifugal force affecting the tracer particles. Between suction and the inner edge of
the wheel, the airflow is very rectified due to the continuity condition. Between the
blades, the radial velocity varies significantly. Based on a mass balance, the radial
velocity at the outer edge of the wheel is about 7 m s
−1 and increases up to ca.
12 m s
−1 .
While at 3000 rpm a relatively uniform velocity profile is established, which follows the above constructed picture, at 9000 and 15,000 rpm the profiles are much less
homogeneous The formation of the profile is due to the secondary vortex discussed
above.
The deflector wheel considered here exhibits an experimental sharpness of cut
of 0.57 at 3000 rpm, 0.59 at 9000 rpm and 0.50 at 15,000 rpm. Comparing these
observations with the flow fields shown before, the reduction of the sharpness of cut
at high revolution rates is not surprising.
This effect will be included in the calculated separation curves shown in the
following.
A parameter which affects the sharpness of cut is the rotational symmetry of the
airflow. At low revolution rates, radial and tangential velocities at the outer edge
of the wheel are very similar and rather low. It is expected that the airflow is more
symmetrical at higher revolution rates where the circumferential velocity of the wheel
dominates. In Fig. 18 the average tangential velocities are shown on the left and the
radial velocities on the right for different revolution rates. The measurements on the
“North” side are indicated in black (cf. vertical plane at radii of 23 and 26 mm shown
in Fig. 15, right) and the “East” side in red (horizontal plane mostly at a radius of
25 mm in Fig. 15, right).
Generally, it is observed that the tangential velocity is more homogeneous than the
radial velocity which was already found for the North side in Fig. 17. At the different
revolution rates, local minima appear which are not at the same positions for North
and East measurements. This indicates that the vortices in the apparatus affect the
flow around the wheel in different ways. The radial velocity, on the other hand,
exhibits a more pronounced and rather systematic divergence. While on the North
side a mostly inwards oriented flow is encountered, the radial velocity on the East
side is nearly zero up to moderate revolution rates. Only at higher revolution rates,
the radial velocity is slightly oriented outwards. This missing rotational symmetry
needs to be taken into account in the following considerations about the particle
separation.