Process Puzzler: Eliminate Evaporator Effect

Readers solve an evaporation pump problem.

We’ve just discovered what appears to be a design flaw during the first hazard and operability study involving solvent-extraction evaporators. The evaporation system is a three-effect backward-feed design using long-tube vertical evaporators. The viscosities should be moderate, i.e., below 100 cP; there’s only a slight possibility of solid build-up unless the bulk temperature falls below 100°F. With backward feed, 150-psig steam enters the first unit as the feed while a dilute concentration of ferric chloride and hydrochloric acid enters the third unit.

The product, a concentrated aqueous solution of ferric chloride, exits the first effect. It’s passed with the evaporated steam from the first unit to the second and then to the third, where it’s condensed and decanted after the final effect. The problem is the level control and pump cavitation. Operating at the design capacity of 50 gpm, the pump can empty the reservoir, i.e., the receiver, below the pump in less than a minute. This problem is most severe during startup. At this stage of the design what can we do to make this process operate more smoothly? Unfortunately, the evaporators, which require expensive materials, already have been ordered.

The simple solution would be to install a 3-in. pad flange level controller on the reservoir and change out the pump starter to a VFD (variable frequency drive), then control the speed of the pump by level.
Ryan Cariveau, project engineer
Archer Daniels Midland, Cedar Rapids, Iowa

Recycle part of the pump output to the inlet of the effect feeding the pump. This would be controlled by the level in the bottoms. If the level gets too low, the recycle could be increased by a recycle control valve. The product would be removed based on a set-point level; should it get low, some material will be recycled until it gets back above the set point.

The feed rate to the high water effect would be controlled by the temperature or density or concentration of the material in the second or third effects. Control will be interesting, but I’m not motivated to figure it out right now.
Bennett Willis, professor
Brazosport College, Lake Jackson, Texas

I think there is a design error. Here are my thoughts:
1. Consider a minimum-flow bypass back to the evaporator or feed to the evaporator.
2. Program a cutoff for the flow rate from the evaporator if the level drops below a certain minimum value.
3. Review the basis of the pump capacity. Is the design capacity from the material balance or is it based on the maximum expected flow? If it is based on the maximum expected flow, it may be possible to trim the pump capacity.
G. C. Shah, HSE project manager
Mustang Engineers, Sunnyvale, Calif.

It sounds like the pump is oversized for your application if it can empty the reservoir in less than a minute. I recommend scaling down the pump size or slowing it down with a variable frequency drive. This will increase the time to empty the reservoir and the level control should be more accurate.

Regarding the pump cavitating, cavitation occurs when vapor pockets form in a liquid flow because of local reduction in pressure. Undissolved air can also cause cavitation where the local total pressure is quite close to the vapor pressure. Cavitation can occur in any machine handling liquid whenever the local static pressure falls below the vapor pressure of the liquid. (This is likely to be a problem with evaporators.)When this occurs, the liquid can locally flash to vapor, forming a vapor cavity and changing the flow pattern (to two-phase flow). The vapor cavity changes the effective shape of the flow passage, thus altering the local pressure field. Since the size and shape of the vapor cavity are influenced by the local pressure field, the flow may become unsteady. The unsteadiness may cause the entire flow to oscillate and the machine to vibrate.

As cavitation commences, it reduces the performance of the pump rapidly. Continued operation can also lead to erosion damage or surface pitting. Cavitation can be avoided if the pressure everywhere in the machine is kept above the vapor pressure of the operating liquid. At constant speed, this requires that a pressure somewhat greater than the vapor pressure of the liquid be maintained at the pump inlet (suction). It is important to carefully limit the pressure drop in the inlet piping system.

A pressure gauge on the suction line before the pump will help you identify the flow condition of the liquid. You can also use this to test at what suction pressure the cavitation occurs and adjust your process accordingly.

There’s a useful web video on cavitation at:

(Since this condition probably occurs during startup and shutdown it may be necessary to run the evaporator in manual mode, not automatic, during these times.)
Errol Williams, chemical process engineer

The solution to the puzzler is a recirculation line. I would pump a portion of the liquid back into the receiver. (All of the liquid should be sent back to the receiver during startup.)
Dr. V. Ravichandran, technical director
Sartime Horological (P) Ltd., Perungudi Chennai, India

Could the level control issue be eliminated if you put a bleeder valve on the pump to get rid of the cavitation problem?
Jim Deary, shift supervisor
Bayer Films, Berlin, Conn.

This is an all too common problem. With the equipment on order, you could be looking at a serious hiccup in your project: shipment delay and unexpected costs for an unanticipated nozzle addition. Or you could recirculate flow to pipe —preferably one going back to the feed. But I am getting ahead of myself. First, you must review the operations manual and the programmable-logic-controller requirements because a new flow control loop will be required for startup and shutdown of the evaporators.

There are three ways to protect the pump from cavitation: flow, pressure and level. The least expensive, since the level is already part of the control, is to use the level control to turn off the pump. I’m assuming that an agitator stirs the bottom of the receiver. Because the impeller is ineffective when there’s too low a liquid level above it, the pump may be needed to provide what little agitation it can; pumps make poor agitators but crystallized solids can kill the whole startup. This probably rules out shutting off the pump.

So, with few options left, we probably need the pump circulating at all times via a recirculation line. We could get fancy with flow meter and flow control valve but, remembering Occam’s razor — i.e., don’t make any more assumptions than the minimum needed, the best way may be to size a restrictive orifice for this purpose.

Size the orifice for twice the pump minimum flow and check its capacity after including this loss; a larger impeller may be needed. Next, install a nozzle in the evaporator feed line because this will have the least impact on the project schedule. Finally, review the entire process for mistakes as this type of error is really serious.
Dirk Willard, senior process engineer
Swenson Technologies, Monee, Ill.

We replaced a thermocouple in a continuous hydrocarbon fluid-bed reactor during a turnaround. It’s located in a duct where high erosive flow occurs because of high velocity catalyst beads. We also replaced the carbon-steel thermowell with a high-nickel-alloy one because of a report of a slight corrosion problem. Since then, the thermocouple has failed every six months when previous thermowells lasted a year until the turnaround. What do you think the cause is and what can we do about it?

Send us your comments, suggestions or solutions for this question by July 10. We’ll include as many of them as possible in the August 2009 issue and all on Send visuals — a sketch is fine. E-mail us at or mail to Process Puzzler, Chemical Processing, 555 W. Pierce Road, Suite 301, Itasca, IL 60143. Fax: (630) 467-1120. Please include your name, title, location and company affiliation in the response.

And, of course, if you have a process problem you’d like to pose to our readers, send it along and we’ll be pleased to consider it for publication.

Join the discussion

We welcome your thoughtful comments. Please comply with our Community rules.
All comments will display your user name.

Want to participate in the discussion?

Register for free

Log in for complete access.


No one has commented on this page yet.

RSS feed for comments on this page | RSS feed for all comments