Process Puzzler: Tell the Bosses the Truth

Management’s goal for a material balance likely isn’t feasible.

Share Print Related RSS
THIS MONTH’S PUZZLER
Our company has been taken over. The new management wants to improve our energy efficiency and accuracy of our material balance (see Figure 1 online at http://www.ChemicalProcessing.com/articles/2009/171.html). They’d like to bring the balance to ±1% accuracy. Several flow meters yield questionable data: 1) an orifice meter measuring a wet gas (A in the figure); a vortex shedding flow meter with a tendency for fouling (B); a meter with a competing recycle line (C); a Coriolis meter with a gas problem (D); a vortex meter with startup difficulties because of gas entrainment (E); a Coriolis meter that measures product successfully (F); and a venturi tube meter that occasionally suffers plugged taps (G). The turndown on the process is about 3:1. How can we reduce our energy consumption and improve our material balance?

YOU CAN'T DO IT
A ±1% material balance accuracy probably is not possible given the realities of this multi-phase process. There’s a weight feeder, which means a solid/gas material balance, a purge gas/wet gas, other wet gas issues and fouling. Coriolis meters, the best hope for achieving high accuracy, can’t deliver when wet gases or slurries are involved. The best accuracy with these troubles is probably in the 10% range. With that said, there are ways to improve accuracy.

The accuracy of a wet-gas orifice plate will be off by 10%–15% without compensating for temperature and pressure. If a local temperature transmitter is available, use a program block in the distributed control system (DCS) with the differential pressure (dP) meter. Better yet buy a multivariable transmitter to replace the dP transmitter and install a Pt thermocouple. A multivariable transmitter compensates for the temperature and pressure drop in the transmitter. Insulate and heat-trace the wet-gas line to avoid surging from two-phase flow.

Vortex shedding meters, which infer flow from velocity, have a wide turndown, up to 10:1, require as much pipe upstream and downstream as a dP orifice meter, but don’t do well with dirty fluids or slurries. Fouling is a problem. I once saw a plugged vortex meter read three times estimated pump output.

Consider some ways to reduce fouling. With steam, a knockout pot is needed to avoid a flow loop spike. The best meter for fouling with a conductive liquid is a magnetic meter. For hopeless cases, use a weigh cell. Nuclear magnetic resonance flow meters seem poised to compete in the non-intrusive arena but I am not aware of any industrial meters available.

As for the recycle line, move the meter past it to get an accurate flow of reactor feed.

Some redundancy is required to check the material balance and permit future improvement projects. Add a flow meter to column recycle lines and product feed to the column. A meter after the reactors might work well, especially if the fluid is a gas. Another meter at the exit stream from the weigh-feeder reactor could be useful as a diagnostic tool.

Now, let’s consider energy conservation. A chilled water cooler and refrigerant cooler point to an exothermic reaction. Why not preheat the feed with products? If product is temperature-sensitive, design a cascade bypass temperature-flow loop. If temperature is not a problem, design a manual bypass.

Talk straight to the new management. Tell them you can do a lot to improve process reliability and energy efficiency but ±1% accuracy for the material balance is impractical.

Dirk Willard, senior process engineer
International Steel Services, Inc., New Caledonia


JANUARY’S PUZZLER
Our old vapor-feed 11-tray bubble-cap fractionating column has developed a problem. The liquid product collected from the reboiler is weak. Reboiler product contaminates the volatile liquid from the condenser. How do we investigate the problem? What are possible causes of contamination? Is there any way to continue to run the column until the next turnaround, a month from now?


Send us your comments, suggestions or solutions for this question by December 11, 2009. We’ll include as many of them as possible in the January 2010 issue and all on ChemicalProcessing.com. Send visuals — a sketch is fine. E-mail us at ProcessPuzzler@putman.net 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.

Share Print Reprints Permissions

What are your comments?

Join the discussion today. Login Here.

Comments

No one has commented on this page yet.

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