Special Section: Flow/Level
Joy of soy flow integration
One green soy plant sees control use for new flowmeters, level indicators
By Tim Foster and Rob Swim
As a $40 million, 195,000 square-foot soy protein isolate (SPI) manufacturing facility in South Sioux City, Neb., Green Planet Farms’ main focus is to serve organic and non-genetically modified protein isolates to food and beverage producers worldwide. Making sure flow and level measurement are accurate and affordable plays a critical part in the process, especially in maintaining its green reputation.
Due to the company’s organic process, its dry soybean protein has a more stable nature—something the market has not previously seen. This means when drinking a protein beverage, the protein does not sink to the bottom of the glass and require frequent shaking or mixing. The protein also has no detectable flavor, a big advantage over other proteins that carry a soy aftertaste. Other SPI manufacturers begin with partially defatted soy flour and rely on hexane (a gasoline byproduct) during the extraction process.
To transform soybeans into healthier, minimally denatured and 100-percent-soluble protein isolates, the company begins with soybeans that have been milled into full-fat soy flour. Its patent-pending, warm-water process, known as G2O, then naturally extracts the protein in the flour, separating it from the fiber, fat, and sugar. This process pumps the elements along through a network of large stainless-steel vessels and then drops the extracted liquid protein into a 130-foot-tall spray dryer. Once the substance dries, the isolate is packaged and sent to food processors for a host of products, such as protein shakes and snack foods.
Maintaining its “green” moniker, the company’s environmental responsibility includes closely controlling processes, such as minimizing wastewater going out of the plant. This requires the use of flow measurement in heat and energy recovery systems—taking all the heat out of process water and putting it into incoming water to avoid heating it. By monitoring a combination of temperature and flow, they know how much of the incoming water they can actually heat. The wastewater treatment plant requires water at a certain temperature, so the SPI plant provides them with what they need and takes the rest out.
The flowmeters see use in two parts of the process, to measure the flow of process water (which is one of the ingredients in the process) and to measure product inside the process for process control. The process is basic and depends on how information is used, such as using flowmeter information with the control system to connect to all ingredients. Flowmeters see use on several lines that combine together and are controlled by the control system. They are more of a control point rather than just a measurement. The information combines together, and all flows are regulated based on flowmeters.
The control valves limit the amount of water and process ingredients in recipe control. The integration gives the plant a fully automated system. Technicians spend most of their time running and getting production samples, which they analyze and use to fine-tune the automated system. With organized product, in the basic term, there are differences in raw materials. They differ in moisture and other paramaters the company selects them for. An automatic system does not know the parameters, so it requires constant tweaking, but after receiving information, the system itself does all the adjustments.
The plant uses diaphragm-type level indicators in water and product tanks because they are accurate and unaffected by chemicals or the plant’s products. Some products are foamy and do not lend themselves well to sonic level sensors. The diaphragm weighs what is in the tank; it senses pressure and ascertains the level based on pressure. Sonic does not read through foam, neither does radar, or not as well. While sonic and infrared have the ability to reach through foam, the diaphragm types are simple and accurate, repeatable, and they work, with little or no maintenance.
The process can be challenging due to the product’s tendency toward stickiness, the foam, as well as a variety of viscosities and densities. But the flowmeters and level indicators are user friendly for calibration. Once calibrated, they work. They do not require multiple calibrations or time spent checking equipment that is supposed to be checking what you are doing. They are time and labor savers. If you can trust your instruments to be accurate, it adds credibility to your process. That means you can spend time optimizing processes rather than just maintaining them.
The product itself is what makes the flow and level processes different. The key is the ease of integration with the control system and the information operators can rely on that comes back from the flow/level equipment. That provides the basis for the company’s whole control system. In fact, the performance of the flow controls was better than anticipated, so smaller ones could have been enough.
The start up of this plant was almost a nonevent, from the standpoint everything worked and operators were able to quickly jump from manual checkout to full automation in the plant. Based on the success of this plant, the company plans to expand its facility and open another organic soy processing plant in 2011.
ABOUT THE AUTHORS
Tim Foster is vice president of engineering at Green Planet Farms in South Sioux City, Neb. (firstname.lastname@example.org). Rob Swim (email@example.com) is global program manager in the Process System Integrator program at Rockwell Automation in Mayfield Heights, Ohio.
Magmeter brushes by microbubbles
By Nathan Chui and Stephen Tardif
When microbubbles in the permeate lines at a water utility in southern Florida caused highly erratic and inaccurate flow measurements, plant managers acted by purchasing new electromagnetic flowmeters (magmeters). As a trial project, the plant installed a new 6-inch electromagnetic flowmeter on total permeate flow and saw pleasing results with steady, accurate, responsive readings, regardless of flow disturbances.
Microbubbles in the permeate make flow measurement and pressure control difficult. Erratic and noisy signals from pulsed DC electromagnetic flowmeters previously tried at the plant degraded flow measurement accuracy and upset the SCADA control system. Adding signal conditioning with long damping times smoothed flowmeter output, but also decreased accuracy and resulted in slow control system response.
This particular utility views flow measurement so important because it controls the pressure that forces incoming water through nanofiltration membranes for purification.
It also monitors the performance of nanofiltration trains (indicating when cleaning or replacement of the membranes becomes necessary) and serves as documentation for state permits that limit the amount of water drawn from the utility’s wells. Flow measurement here acts as a check on revenue from downstream customer meters used for custody transfer.
The water treatment plant withdraws groundwater from the Biscayne Aquifer via four wells. The plant processes the water, removing sediments, harmful bacteria, and certain minerals. The water is disinfected by chlorination and fluoridated before entering the distribution system.
The plant processes water through five nanofiltration trains, producing 8 million gallons per day of purified potable water. The plant initially injects de-scaling chemicals into the well water. The water then passes through pre-filters that remove sediment and other solids and particulates.
Water from the pre-filters enters a dual-stage nanofiltration system. It first passes though a pump with a variable frequency drive. The pump increases system pressure to force the water through both nanofiltration stages. Each stage consists of 42 nanofiltration membranes connected in serial tubes about 20 feet long. Concentrate from the first stage goes through a second set of membranes in the next stage before being sent to injection wells.
The performance of the nanofiltration membranes degrades over time. As performance degrades, total permeate flow through the final flowmeter begins to decrease. The flowmeter’s output signal connects to a SCADA control system, which boosts VFD pump speed to increase the pressure driving the permeate through the nanofilters. Pressure cannot be increased indefinitely, however, and at some point the plant must clean or replace the nanofiltration membranes.
The low-level signal from the primary magmeter goes to a transmitter with a local readout of flow rate for operators. The transmitter sends a 4-20 mA signal representing flow to the digital SCADA system. SCADA calculates an appropriate control signal to adjust system pressure by increasing the speed of the VFD-driven pump.
The new magmeters combined the advantages of AC and DC excitation and offered a steady, accurate, and responsive flow reading, regardless of flow disturbances from microbubbles in the permeate lines. It also eliminated the zero shift usually associated with AC magmeters.
ABOUT THE AUTHORS
Nathan Chui is a project manager at Alpha Valve & Controls in Tampa, Fla. (Nathan@alpha-controls.com). Stephen Tardif is a senior applications support engineer at ABB in Warminster, Penn. (Stephen.firstname.lastname@example.org).
Return to Previous Page