(79b) Feed Entry into Parting Boxes: What Works and What Does Not

Achieving good liquid distribution is crucial for the performance of packed towers and multi-pass tray towers. Amazingly, one device that has received little attention in the published literature is the distributor parting box. While good-practice guidelines are available, the published literature is short of work on parting box hydraulics and tested guidelines.

To gain insight into the parting box hydraulic, we built a scaled-down model, geometrically similar to the parting box in a hydrocarbon tower, and tested it with water. This model provided generic insight into parting box hydraulics as tested in a realistic parting box design.

Tests videos that we will show indicate that when feeding from a sparger pipe, the liquid in the test parting box was highly frothed up due to the water jets from the pipe entraining air into the water upon impact. At high rates this generated excessive froth heights in the parting box, and premature overflow. The liquid split to the five clusters of holes at the floor of the parting box was good at the maximum rates but quickly deteriorated as the liquid rate was turned down.

Adding a wall-to-wall Perforated Impingement Plate (PIP) with either 20% or 10% open area, mounted either 2” or 4” above the parting box floor gave a very good split of liquid to the clusters of holes at the floor of the parting box both at the maximum rates and at turndown. However, an air gap formed beneath the PIP, with tall froth above, producing premature overflows at high rates.

We also tested a common design in which dip tubes instead of a sparger pipe are used to bring the incoming liquid 2” above the floor of the parting box. This eliminated the frothing in all but the lowest rate, but the liquid heights in the parting box were considerably higher than those calculated from the orifice equation, and the split of liquid to the floor holes was mediocre. We theorized that excessive horizontal momentum at the parting box floor caused these issues.

Our tests identified a few satisfactory innovative methods of feeding the parting box. These include using dip pipes fitted with tees that directed the liquid towards the parting box walls; using a Vented Impingement Plate (VIP) with a venting area 40-50% of the box cross section; and using guide tubes with perforated tips of 20-40% open areas. Each of these eliminated the excessive liquid heights, gave good split of liquid to the clusters of holes in the floor of the parting box, and a head-flow relationship closely matching the orifice equation.