• Ei tuloksia

9. SINGLE PHASE MODEL SIMULATION RESULTS

9.4 Outlet configurations

The purpose of the Outlet study was to evaluate the effect of changing gas outlet configuration to fluid flow mainly around the demister area. Three different outlet configurations were simulated: an outlet with rounded edges, a round solid plate placed below the outlet and a square perforated plate placed below the outlet. According to generally used friction loss charts, rounding the outlet with an r/D ratio of 0.15 reduces the friction loss coefficient K by 92% from 0.50 to 0.04 as compared to a sharp edged design.

Sketches of the mentioned modifications are presented graphically in Fig. 51.

FIGURE 51. Schematic drawings of the three different configurations studied in the Outlet study. Roundness of the outlet exaggerated for illustrative purposes.

The configuration of the holes in the Perforated plate case is presented in Fig. 52.

FIGURE 52. Positioning of the holes in millimeters in the Perforated plate case

Only the Impact Plate Type 1 distributor and the high capacity design values were used in modeling of the cases in the Outlet study. Maximum and standard deviation values of vertical velocity across a plane 5 cm below the demister pad are presented in Figs. 53 and 54, respectively.

FIGURE 53. Maximum values for vertical velocity 5 cm below the demister pad in m/s in three modified Outlet configuration -cases with high design capacity. Base geometry case

results presented as reference.

FIGURE 54. Standard deviation of vertical velocity 5 cm below the demister pad in m/s in three modified Outlet configuration -cases with high design capacity. Base geometry case

results presented as reference.

Figs. 53 and 54 indicate that the Rounded outlet has the biggest impact on fluid flow among the Outlet configurations. Rounding the outlet increases the fluid maximum velocity at the demister inlet slightly as seen in Fig. 53. This is due to absence of resistance caused by sharp angles to the outlet fluid flow. The uniformity of the flow profile at the demister inlet is increased with the Rounded outlet as seen in Fig. 54. Even though the peak velocities can increase slightly, overall the Rounded outlet produces a smoother flow profile at the demister inlet compared to other Outlet configurations. Based on basic fluid

0

Base Geometry Rounded Outlet Solid Plate Perforated Plate

U_Z, max [m/s]

Base Geometry Rounded Outlet Solid Plate Perforated Plate

U_Z, std [m/s]

dynamics principles, the effect of the Rounded outlet should be more pronounced with liquid flows, since they experience higher friction forces at sharp corners due to higher viscosities. Close up views of the vertical velocities in the outlet area on a vertical plane for each Outlet configuration are presented in Fig. 55.

FIGURE 55. Vertical velocity profile on a vertical plane in each Outlet configuration case.

Standard deviation value calculated across demister inlet. Refer to Fig. 26 for plane location. High design capacity, Impact Plate Type 1 distributor.

Fig. 55 confirms that the solid and perforated plates function as expected. It also confirms that the Rounded outlet enhances flow through the outlet. This is seen as reduced flow velocity outside of the demister frame. A summary of vertical velocity planes 5 cm below the demister pad for each Outlet configuration case is presented in Fig. 56.

FIGURE 56. Vertical velocity profile 5 cm below the demister inlet in m/s in each modified Outlet configuration case. Base case presented as reference. High design

capacity, Impact Plate Type 1 distributor.

Visually, all of the flow profiles in Fig. 56 exhibit acceptable evenness. The Rounded outlet stands out with a slightly lower standard deviation among the Outlet configurations despite the formation of a small high velocity zone near the back wall of the vessel. Based on this, it can be deduced that the low standard deviation value is the result of the whole flow field obtaining velocities very close to the mean velocity at the demister inlet with the exception of the high velocity zone. In actual operation of the separator, this type of uniformity does not yield any additional benefits as long as the maximum velocities remain below the acceptable limit. To more closely study the effects of the outlet configurations, vertical velocity planes were also extracted at two locations inside the demister frame. The lower plane was extracted 5 cm above the demister pad and the upper 5 cm below the plate. These planes are presented in Fig. 57 for all of the Outlet study cases with only the area inside the demister frame included in the pictures.

FIGURE 57. Vertical velocity on two planes inside the demister frame in the Outlet study cases. Only area inside demister frame is presented. Top picture indicates plane locations (in red) between demister and the plate (in blue). High design values. NOTE! Different

color scale compared to other velocity figures.

Fig. 57 indicates that the rounded outlet does not significantly change the flow profile flooding similarly than as a result of upstream velocity increase. In the calculated cases, the maximum velocities downstream of the demister remain at 3 m/s which is the same as for the Base geometry. Therefore it is unlikely that any of these modifications will contribute to flooding of the demister.

Overall, it can be concluded that none of the Outlet configurations can be declared best when employing the well performing Impact plate distributor. Further studies should be conducted to determine the possibilities of improving flow with outlet modifications when using distributors with worse performance characteristics.