This Month’s Puzzler
Our research team has changed the flow characteristics of a product. By my measurement, the viscosities of the ingredients now are in the 100-cP range, while previously 30-cP was the maximum. The researchers still are monkeying around with the batch mix schedule, so I don’t have a final value on the flow properties.
Thankfully, we use gear pumps for ingredients. The product ran about 20 cP, which allowed us to rely on our old 80-gal/min magnetic-drive centrifugal pump to send the product, which is a weak slurry, to a filter press. The total dynamic head is 148-ft-WC (about 64 psi). This pump is easily cleaned-in-place. Space around the 3,000-gal reactor is very tight, preventing me from installing a new pump off the suction line of the current pump. I also am concerned about the agitation. I have to be ready to change this reactor to the new product within a few weeks. What are your thoughts on the product delivery pump and reactor? Can we successfully make the product?
Tune The Process First
Because the problem statement lacks sufficient information about the reactor, potential variation in the product viscosity, and equipment spacing at the site, I only can offer general suggestions. Consider the following issues:
1. Typically, centrifugal pumps are not a proper choice for reliable operation with liquids above 100 cP. For moderate to high viscosities, choices include gear, lobe, piston and progressive cavity pumps. Of course, each of these has pros and cons. For instance, liquids with abrasive particulates would cause wear of gear or piston pumps. If you anticipate wide variations in product viscosity, a progressive cavity pump might be a good option.
2. Because space is limited in the immediate vicinity of the reactor, the suction line to the new pump would be longer. This could cause low suction head at the pump, which, in turn, would be problematic for a piston or progressive cavity pump. A gear pump, on the other hand, is relatively less sensitive to low suction head. In the event of low suction head, see if you can increase pressure in the reactor without affecting product quality.
3. Take into account the flow variability of the product. If you anticipate large variations in the product flow, sustained operation at low flow rates will cause excessive wear of a progressive cavity pump. So, at low flows, consider spillback for such a pump.
4. If you do not expect product viscosity to go significantly above 100 cP, an option worth thinking about is a turbine agitator with high turbulence. Typically, mixing liquids above 600 cP calls for laminar impellers. Because the mixing energy dissipates a short distance from the edge of the blades, these blades would need to extend to a good part of the diameter of the vessel. Also, consider variable speed.
5. In view of the multitude of tasks you need to accomplish in a relatively tight time schedule, get outside help early, including from mixer and pump vendors. Keep the project group and operations and maintenance folks posted on the progress and problems.
6. Note: A CP column “Pumps: Get into the Thick of Things,” by Dirk Willard offers practical pointers on the qualitative aspect of selecting pump types for the viscosity spectrum of liquids.
GC Shah, consultant
Houston
Think Radically
Given how viscosity increases the horsepower requirements of centrifugal pumps, you have two options: raise the product temperature or dilute the product. Dilution hurts the production rate and increases the utility cost in drying. Besides, it might negatively affect the system downstream from the heat exchanger; assessing this could involve a heat-and-energy balance of the entire process — upstream of the exchanger to the final product.
Raising the temperature might squeak you by if the pump motor suffices. Get some viscosity data at temperature, then discuss these data with the pump vendor, and run a check against viscosity correction curves. You might be alright. If not, you might try air dilution to thin the liquid even more or a booster pump. An air-driven trash pump is more tolerant and might serve as a temporary patch for the problem of moving the product by bypassing the old pump. One word of caution on introducing air to a liquid stream: you gradually will damage the pump no matter how carefully you introduce the air.
Next, let’s turn to the agitator. I wonder if the people in research considered this. If you increase viscosity by a factor of three, reaction times will likely lengthen unless you take drastic steps. I assume you have budget constraints; money blows most problems out of the water. Going with my assumption: 1) maximize the shaft speed; 2) if possible, premix ingredients before adding to the reactor, especially solids but also liquids; 3) decrease the reactor temperature to delay the reaction until mixing is complete and then heat up to pump product; 4) dilute to reduce viscosity and promote mixing.
As usual, R&D boxed you into a corner. This is why engineering has to be integrated into product development. Good luck. If you escape, you’re a hero.
Dirk Willard, consultant
Wooster, Ohio
June’s Puzzler
We make an ethoxy compound using a water-based process. Our evaporator is supposed to operate continuously but instead the operators run it in manual — with a lot of baby-sitting. The operators complain the level switches and level transmitter don’t work as advertised. So, instead, they judge level by looking through three sight gauges. This requires distinguishing wet solids from mere splatter; new operators must learn the difference between actual level and a dirty glass. The design engineer somehow thought a 2-in. pipe made sense for pumping the product, a wet slurry.
I looked at the material balance. Continuous processing is possible, if the flow rate were 10 gal/min through a pipe where the solids settle out at 4 ft/sec — and with working level measurements! Obviously, a 2-in. pipe forces us to dump the product into the dryers in loads measured by estimating the level using the sight gauges.
Temperature control is nearly impossible. Measurements by a resistance temperature detector vary as much as 10°F from sample temperatures and readings with infrared guns.
What can we do to make this a continuous evaporation train? Should we do this? Is there anything we can do to make this process less labor-intensive as a batch process?
Send us your comments, suggestions or solutions for this question by May 13, 2022. We’ll include as many of them as possible in the June 2022 issue and all on ChemicalProcessing.com. Send visuals — a sketch is fine. E-mail us at [email protected] or mail to Process Puzzler, Chemical Processing, 1501 E. Woodfield Rd., Suite 400N, Schaumburg, IL 60173. 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.
