Sensors and other devices are the foot soldiers in automation architecture, but most food and beverage manufacturers still are looking for full value.
You know an equipment category has arrived when it starts getting its own industry-specific center. For the food production sensors that do the heavy lifting in instrumentation technology, that day could arrive earlier rather than later.
A consortium of Ohio food companies, technology providers and academics affiliated with Ohio State University’s food science department is seeking state funding to underwrite the Food Industry Sensor Center, a resource for food and beverage companies looking for both off-the-shelf and new or customized sensors to help them move away from lab testing to on-line sampling or in-line monitoring to correct process variability faster. Economics is the driver: Whether the goal is reduced energy use or less raw material waste and rework, manufacturers are realizing quick paybacks by eliminating waste and inefficiencies that used to cost much less and were accepted as part of doing business.
The sensor project is an offshoot of the Center for Innovative Food Technology, which recently conducted a study of instrumentation needs and trends. Chemical sensing and biosensors are among the most in-demand devices, but the greatest need is for a more fundamental measure: level sensing.
Level measurement ranks with pressure, temperature and flow as the most prosaic tasks performed by instrumentation devices, and a number of technologies are deployed to deliver that information to process managers. Unfortunately, notes Rockwell Automation’s Fred Discenzo, their information needs are not well met. Determining product level becomes particularly challenging when product is submerged and foam is present, the manager of diagnostics & sensors at the Rockwell’s Advanced Technology Laboratories in Mayfield Heights, OH says. Current options are imprecise or require frequent maintenance. Nanoscale MEMS devices (microelectromechanical systems) are promising, though development work will be needed to transfer the MEMS used in automotive and other sectors to food and beverage applications.
Discenzo’s lab gets requests from all types of manufacturers for technology requiring R&D before deployment. “Most of the requests are from the food industry,” he confides, “and most of those are for sensors.”
Envelope-pushing sensors are exciting, though many suppliers argue that manufacturers are not leveraging the capabilities already built into their instruments. “Most of the beverage and water quality analyzers have diagnostic and communications capabilities, but PLCs and other controls have not been set up to take advantage,” says John Martin, food & beverage marketing with Irvine, CA-based Emerson Rosemount Analytical. The situation is improving in some plants, provided digital devices are deployed. An estimated 20 million HART devices are in the field, and for those manufacturers, “you can strip out the HART information from the process parameters and use it free of charge,” Martin says.
Need-to-know information about process variation also is being lost, with mass flow meters the poster child example. Unlike one-dimensional devices, mass flow meters go well beyond volume to correlate information on mass, Brix and even temperature. Multi-tasking also is possible to a degree with devices that measure phase changes from liquid to gas. Unfortunately, these devices usually are installed by OEMs who deliver a skid solution, and unless the manufacturer specifies the added capabilities in the RFP, suppliers shy away from using high-end instrumentation. “When the OEM is focused on shipping the skid and the end-user is focused on lifecycle cost,” says Ron Salerno, business manager for flow, level, pressure and temperature at Spring House, PA-based Siemens Energy & Automation, “value can get lost if it gets into a price discussion.”
Emerson’s Martin agrees, adding, “OEMs and integrators are cranking out what the customers need, but they’re not fully utilizing the diagnostic capabilities of the devices. For example, a coriolis meter can identify incremental density changes that signal a problem with entrained air, which would be a good thing to know.”
ABB Instrumentation in Ambler, PA struggled with the instrumentation knowledge gap a few years ago when it upgraded its magnetic flow meter for food. The new meter represented a quantum improvement in sanitary design and signal output, and although integrators recognized the value, end-users only saw a higher cost. “Food and beverage is stuck right now,” suggests Isabelle Durso, an ABB product manager. “It’s very difficult to reach the end-user, and the end-user isn’t keeping up with the technology.”
The good news for manufacturers with little or no instrumentation investment is that they are not tethered to the 4-20 milliamp data transmission that characterizes the analog world. Like developing countries that leapfrogged land lines to cell phone communication, those companies don’t have to rationalize the expense of rewiring their plants for digital devices. Instead, they can install a bus system such as Foundation Fieldbus or Profibus or, even better, take advantage of wireless protocols.
The low communication power required in the recently adopted Wireless HART standard should make battery-powered sensors feasible in plants, though the impact in food and beverage may be years away (see related story on page 90). For the dairy segment, wireless is not an option. “The PMO (Pasteurized Milk Ordinance) requirements prohibit the use of Foundation Fieldbus or Wireless HART protocols,” points out ABB’s Durso. Device settings in pasteurizers and other process equipment are sealed by public health inspectors, precluding remote changes by a wireless device.
Where regulations aren’t an issue, a closer look at wireless devices would be a positive change, believes Craig McIntyre, an industry manager at Endress+Hauser Inc., Greenwood, IN. “Wireless is like sushi: Most people who don’t like it haven’t tried it,” he says. “If you’re trying to measure temperature at a point 50 ft. up, putting in a wireless sensor makes a lot of sense.”
Sophisticated automation software is only as good as the sensors feeding data to it. Rockwell recognized its sensor line wouldn’t be up to par with the new possibilities for in-line instruments, and it addressed that shortcoming by forming an interoperability collaboration with Endress+Hauser. Pretesting with software solutions should enable E+H devices to integrate more quickly and reliably. This, in turn, will help expand applications to monitoring of specific gravity, oxygen content, pH and other variables that can affect product quality, says Charley Rastle, Rockwell’s food & beverage point man.
Those types of relationships help address the remote monitoring and calibration abilities that often have eluded manufacturers. Instrument calibration is uneven from one plant to another, depending on the process in place, points out Ola Wesstrom, E+H’s food & beverage industry manager. “Some do a great job, but it’s not as easy as it looks,” he says. “The proverbial bucket test is only as good as the person filling the bucket.” Automated calibration could pay handsomely in reduced product giveaway and waste.
The precision of the sensors themselves is continuing an upward trend, with the quality of digital signal processing following the same curve as electronics in general. “Even coriolis mass flow is getting better,” Wesstrom says. “The standard used to be 0.1% accuracy; now it’s pushing 0.05%.” Similarly, hydrostatic level sensors used to be accurate to within 0.2-0.5%. Today, “all of them are better than 0.1%,” he says.
Breweries constitute the leading industry segment in terms of instrument deployment, with the mashing and fermentation steps particularly fertile ground. Zero tolerance for oxygen presents multiple opportunities for in-line sensors, and detection of naturally occurring diactyl during fermentation is being addressed with nanosensors such as field asymmetric ion mobility spectrometry (see “Food’s chemical signature,” Food Engineering, August 2008). Maintenance can be a drag on deployment, as some brewers have found with pH sensors used to maintain separation between the wort and CIP water. Citing a Belgian brewer that was dealing with sensor failure every three days because of plugging and coating, Emerson technicians say new electrode designs and high-heat glass formulations in pH sensors developed in recent years can prevent plugging and coating for up to three months. Servicing with a gel electrolyte extended the Belgian brewery’s sensor life an additional three months.
Another sensor advance designed around CIP monitoring involves the Rosemount four-electrode conductivity sensor, which eliminates the need for two sensors in dairy and other applications because of its dynamic range. Cleaning solutions have high conductivity, while rinse water has very low conductivity. Processors eager to resume production and reduce the amount of wastewater going through the line had to use two sensors. The new unit meets European EDHEDGE sanitary standards, with 3A approval expected by early autumn.
While dairy and beverage products are more conducive to in-line monitoring, instruments increasingly are being deployed for pumpable foods and solids. One of the more promising developments involves guided microwave spectroscopy (GMS). Thermo Fisher Scientific has been nurturing GMS for several years, applying it to moisture, protein and fat measurements in ground beef, masa and other foods that typically exhibit variations in sample depth. For example, the surface of dough used to make corn tortillas dries out quickly, and near-infrared spectrometry (NIR) does not accurately record moisture content below the surface. Likewise, the fat in ground beef moving through a pipe is unlikely to be homogeneous, and alternatives to GMS are unlikely to provide valid measurements.
“We’ve studied GMS head-to-head with X-ray, and it does a better measurement of fat and moisture content,” according to Rick Cash, marketing technology manager-process instruments at Minneapolis-based Thermo Fisher. Unlike systems that extrapolate two frequencies to come up with a value, GMS uses a single wave transmitted across a waveguide chamber (basically, a bulge in a process line) to determine fat and moisture content. The limiting factor is the need to excite molecules in the food: Ice absorbs wave energy and looks the same as fat, explains Cash, so GMS is not effective with frozen product.
Based on the commercial applications of recent years, Thermo is considering dropping some NIR sensors. A reengineered GMS system that is easier to install and operate may accelerate those plans. “It was kind of ahead of its time a few years ago,” says Cash, “but we’re making it very user-friendly now.”
The same could be said about weight-monitoring devices. San Diego-based Hardy Instruments has taken calibration out of the equation with all-electronic scales. “With cells that only move 1mm, these systems don’t need to be calibrated; they just need to be verified,” says Ted Kopczynski, product manager-process weighing. A fail-safe concept involving a redundant memory chip also should make life easier for users who experience a system failure a week after installation. The extra chip is a common secure-digital (SD) card that records parameter changes. Identical to the memory cards found in digital cameras and other portable devices, the SD card can plug into a PC to allow remote diagnostics and repair via e-mail.
The strain-gauge technology at the heart of load cells has not changed in decades, but mounting hardware and methods of assembly distinguish the best units from commodity scales, according to Kopczynski. Premium devices are filled with inert gas to flush atmosphere before sealing, resulting in “zero shifts over time as a result of corrosion,” he says. And mounting hardware that adjusts to expansions and contractions short-circuit the mechanical issues responsible for the overwhelming majority of weighing errors.
Inert gas behind a stainless-steel membrane adds to the reliability and robustness of pressure sensors manufactured by Anderson Instrument Company Inc., Fultonville, NY. Dairy is the firm’s primary focus, followed by beer and soft drinks and then pumpable foods. Rugged design and built-in redundancy are top design considerations. Lower-cost pressure sensors with ceramic diaphragms have become available in recent years, but dairies rejected them because of the crevices they provide for bacterial harborage and the porous nature of the materials, recalls Bill Wilson, vice president-sales & marketing. But sanitary standards are not quite as high in the beverage sector, and a German firm acquired by Anderson three years ago is using dry capacitive technology to deliver ceramic pressure sensors to those manufacturers.
The company recently developed a digital temperature gauge with redundant temperature elements for retort applications, then spearheaded an effort to amend the Federal Code of Regulations to allow the devices as a replacement for mercury-in-glass thermometers. Glass and heavy metal would be unthinkable in most food plants, though the in-container sterilization that occurs in a retort safeguards product. Nonetheless, accuracy to within a tenth of a degree and the elimination of lost batches when a mercury thermometer breaks should win converts to the new technology, Wilson predicts. “You’ll have some push-back from some of the older guys,” but he expects the digital gauges to become standard within a few years.
Energy management is front and center in manufacturing today, and that is helping companies rationalize the installation of new devices. Siemens developed a pneumatic valve positioner that is virtually bleedless. When a typical valve positioner is in steady state, the hissing air is audible. In-house studies suggest that each PS2 positioner installed cuts compressor operating cost $500 a year, assuming electrical costs at 5 cents per kW. With prices higher than that in many parts of the country, Siemens’ Salerno says the savings can be dramatic, not just in terms of energy consumption but also in compressor sizing and the opportunity to defer additional units to meet demand.
“We supply all industries, including off-shore oil drilling, but food and beverage has the worst operating environment of them all,” concludes E+H’s Wesstrom, to which other technology providers say, “Amen.” But the industry offers enormous potential, and suppliers are improving the durability of their devices to withstand temperature extremes, caustic washdowns and other challenges in food plant use. As the precision and reliability of instruments improve, manufacturers will be better able to leverage the benefits they bring to the automated plant.