Basis: Fp-56

Prtasure drop measured in Inches H20/ft

Figure 8.19 (Continued) The latest version of the GPDC pressure drop correlation, (d) Superimposing experimental pressure drop data for a given packing generates a GPDC interpolation chart for this packing.

mental pressure drop data (60). Also, for some packings, the dependence of pressure drop on vapor and liquid loads was not adequately predicted by the Eckert correlation (15,60).

The Kister and Gilt generalized pressure drop correlation for structured packings [GPDC (SP)] charts. The latter limitation above severely obstructed attempts to apply the Eckert correlation to structured packings. Specifically, in structured packings, pressure drop rise with flow parameter is much steeper than the Eckert GPDC curves predict The Eckert curves are based on random packing data, and for these, the capacity fall-off and the pressure drop rise with higher flow parameters are much milder than in structured packings (see Sec. 8.1.10).

Kister and Gill (316) modified the Eckert chart curves empirically to fit structured packing data. For structured packings, this modified chart (Fig. 8.19c) was shown (316) to give much better predictions than the Eckert chart. For flow parameters between 0.02 and 0.2 for nonaqueous systems or 0.01 and 1 for air-water, and for packing factors between 6 and 30, Fig. 8.19c was shown (316) to work well for all structured packings. Packing factors to be used with the GPDC (SP) chart (Fig. 8.19c) are listed in Table 10.1.

Outside the above range, there are regions where Fig. 8.19c gives less reliable predictions of pressure drop. For some structured packings with high (> 30) packing factors, the rate of rise of pressure drop with flow parameter or vapor velocity differs from that predicted by Fig. 8.19c. For nonaqueous systems, a region of prediction uncertainty exists at high flow parameters (F]v >0.3).

The Robbins correlation (89). As an alternative to the GPDC, Leva (70) derived an equation for pressure drop. For a dry bed, pressure drop can be calculated from a momentum balance.

In order to allow for irrigation, Leva modified the dry pressure drop correlation to give

Leva's equation applies to gas-continuous operation only (70,84). At very high liquid loads, pressure drop is not proportional to the square of the vapor velocity (Sec. 8.2.1, Fig. 8.15), and Eq. (8.12) does not apply. According to Billet and Schultes (78,79), the transition from gas-to liquid-continuous operation occurs at a flow parameter [Eq. (8.3)] of 0.4.

Robbins points out that with dry beds and at low liquid loads, liquid physical properties have practically no effect on pressure drop. This is correctly predicted by Leva's equation (8.12), but not by the GPDC chart, because the chart uses liquid viscosity and liquid density.

Robbins found that the constant Cx in Eq. (8.12) correlates directly with the packing factor. This observation permitted him to derive packing factors from dry pressure measurements [applying Eq. (8.12) with uL = 0]. He also found that the constant C2 in Eq. (8.12) correlates well with the square root of the packing factor and the liquid viscosity to the 0.1 power. These findings permitted Robbins to express the curves shown in Fig. 8.15 in a generalized form, giving the equation

where

Was this article helpful?

## Post a comment