Perspectives: Plant InSites

Ease Measurement of Column Internal Flow

Opting for a pump instead of gravity to return liquid to a tower often makes sense

By Andrew Sloley, Contributing Editor

Last month, we looked at using a gravity-flow loop to measure liquid flow inside a column (“Get Some Inside Information”). An often-attractive alternative uses a pump to return the total liquid draw taken from the column. The important engineering fundamentals of this system lie in the pump and control logic.

Figure 1 shows two possible configurations for measuring internal liquid flow in a column using a pump. The upper drawing represents the conventional configuration, i.e., a fixed-speed pump with a flow controller; the lower one depicts a setup with the pump controlled by a variable frequency drive (VFD). The conventional configuration may include a low-flow bypass if rates below the lower operating limit of the pump are expected. Most applications will benefit from a more-modern configuration with level controlled by a VFD.

Total frictional losses in most systems will be low. Typically such losses —from the flow instrument, piping and the return liquid distributor inside the tower — account for at least half of the total system loss. The static head loss between the draw point and return point is relatively small. This makes a good application for a VFD on the pump.

A pump should not run dry. So, control of the upstream level is necessary to protect the pump. This requires adding a level controller span on the collector tray; for effective control the tray must provide sufficient level range, which usually results in high liquid inventory inside the tower. The high liquid level on the tray would make even a small gap on a bolted tray leak a large amount of liquid —undermining the accuracy of measurements because the leaking liquid doesn’t go through the flow meter. So, the tray must be fully welded for leak-free operation.

The control configurations shown assume a centrifugal pump, which means discharge pressure and flow rate are closely linked. Use of a positive displacement pump would require different control configurations.

The pumped system has both advantages and disadvantages compared to a gravity-flow system.

Advantages include:
• The distance between the draw and return points isn’t critical.
• Pressure drop across the flow instrument isn’t a major issue.
• As long as sufficient pipe length for flow conditioning is available, there’s more flexibility in picking instrument locations.
• In vacuum systems, the pump supplies pressure for a sampling station (as shown in the bottom drawing in Figure 1).
• The pump also provides pressure drop for closed sampling stations, if needed.

Disadvantages include:
• The system costs more. The installation requires pump(s), foundation(s), more instrumentation and more-complex control.
• Auxiliary equipment, such as strainers, may be needed to protect the return-liquid distributor inside the tower.
• The critical distance H1 remains. It is necessary for providing sufficient net positive suction head to the pump.
•The pump and associated valves create new locations for potential leaks.
• Unit startup may be more difficult. This often is true when the draw is from a tower with trays and a full collector tray isn’t used to ensure liquid flow to the pump.

Different process requirements may favor either the pumped or the gravity-flow system. However, pumped systems tend to work more often. That’s not due to any magic. Rather, because a pump raises costs, everyone pays greater attention to system fundamentals and layout. As a result, more pumped systems get done right. However, keeping track of the fundamentals and making sure they’re applied correctly always should be a concern.


ANDREW SLOLEY is a Chemical Processing contributing editor. You can e-mail him at ASloley

More from this perspective...


Tackle Tough Services for Centrifugal Pumps with Care


Don't Let Heater Control Get You Steamed