Servos Find a Packaging Niche

May 5, 2005
Flexibility is driving more manufacturers to servo-driven packaging lines, but concerns remain about validation and the “blue screen of death” with PC-based controls.
Although most pharmaceutical manufacturers keep quiet about their new equipment, servo-powered packaging systems are finally claiming their space on the factory floor. While these flexible machines have become a mainstay among consumer products companies, regulatory requirements have presented formidable barriers to their use in pharmaceuticals."The pharmaceutical industry is just getting into seeing the benefits of the servo world — the easy changeovers that make it economical to do short runs and small lots," says David Blauw, electrical engineering supervisor at Bosch Packaging Technology (Minneapolis), which has built several servo-controlled machines for drug makers. "Instead of taking four to six hours to clean, autoclave, and change cams and parts, you can simply reprogram the machine, load up a different type or size of vial and be up and running in minutes." Servo systems are also simpler, cleaner, less prone to breakdown, and easier to clean, Blauw notes.The problem is that servo systems typically rely on PC-based control systems. They present far more validation issues than the PLC controls and mechanical linkages of conventional shaft-driven packaging systems.Yet many pharmaceutical firms are willing to go the extra mile to put servo machines on the floor, especially as they commercialize more specialty drugs requiring varied packing options, Blauw explains. "With cost pressures from competitors, insurers, and large buyers, pharma companies are looking for machines that give them more flexibility," he says.An electronic system does that far better than a mechanical system, Blauw adds. “Pharmaceutical companies may be hesitant to change from the mechanical realm where they're dealing with a known entity, but their production needs are forcing them to do it," he concludes.

An Allen-Bradley MP-Series servo. Courtesy of Rockwell Automation.

Servo advantagesBusiness pressures aside, servo-based equipment offers advantages over shaft-driven equipment, says Mike Wagner, business development manager at Rockwell Automation (Milwaukee). "Before servos, packaging machines had one large AC motor running at a fixed speed," he says. "Mechanical linkages off the main shaft would control each packaging process, translating shaft motion from one speed to another or from rotational to linear motion."For each revolution of the main shaft, you had one product coming out the other end,” Wagner continues. “Every time you wanted to go from vials to bottles or change the size or shape of the package, you had to reconstruct the machine."Servos change the game. Instead of mechanical linkages to a main shaft, each servo (which includes motor, gear head, and software) operates independently."Everything can be modified — pressure, velocity, position," says Wagner. "Before, it would take seven hours to make those changes, so you didn't want to do it unless you had a large enough lot to run effectively. Now you can make the changeover in three to five minutes. It's like loading a recipe."Freeing servos from the drive shaft enables them to adjust independently to each event during the packaging process. "Imagine a simple low- to moderate-speed bottling line," posits Bob Hartwig, vice president of Pester USA, a subsidiary of the German packaging equipment firm. "It's made up of a host of simple machines, all of them interdependent and working in concert. As the system collects data, the servos on the line can adjust independently to any production problem as it occurs."Servos also offer an entirely new level of speed and precision, adds John Kowal, global marketing manager for Germany's ELAU, Inc., Europe's leading manufacturer of servos for packaging. Kowal estimates that 70 to 80% of the stretch banders, overwrappers, case packers and palletizers now sold to pharmceutical companies use servos."One of the great things about servos is their closed-loop feedback,” he says. “Our servo-based capper controls torque to +/- 0.02% versus 20% for a standard clutch-drive capper. We know exactly what values are. We can actually store the torque value for each cap, so if there's a recall, we may only have to recall 1,000 bottles instead of one million bottles." The same data can also be analyzed for trends that indicate wear or a need for maintenance.Since servos use much smaller motors, maintenance is rare. "Servos have a mean time between failure (MTBF) of 200,000 hours," Wagner says.21st century manufacturingFinally, servos are far more affordable than in the past. "Five years ago, servos cost about $10,000 a pop," says Hartwig. "Nowadays, they cost about $5,000 each and prices are still falling." On an inexpensive machine that costs $100,000, adding 10 servos would increase the price by 50%.Add those same 10 servos to a larger unit costing $500,000 to $1 million and the cost is hardly noticeable. The payoff, however, is huge: a large, fast line that is flexible enough to package solids and liquids in a variety of containers in short and long production runs."They're faster, more accurate, easier to control, easier to maintain, easier to implement, and longer lasting," says a project manager at a leading pharmaceutical manufacturer who did not want to be identified. He says his company is installing all-servo packaging lines when old systems wear out and when it builds new plants. "We're in the 21st century, and this is a new way to run plant. Servos help us predict problems before they occur. They tell us where to troubleshoot. Everything about them is easier. They are a step above driveshaft machines," he says.Blue screen of deathThat's the good news. And if it were the only news, then every pharmaceutical company would be using servos. That is not yet the case, and the reason has to do with servo system controls.To appreciate why servos can make the hair stand up on the back of a manufacturing engineer's neck, look first at the old-fashioned drive-shaft packaging lines. They may not be flexible or precise, but their mechanical linkages make them very, very predictable. A PLC using ladder logic engages one function after another. It is a system that is relatively easy to validate.Servos, on the other hand, are independently operated modules that depend on software rather than gears, chains, and cams to execute their operations. Many of them use PC-based processors. This is both a strength and a weakness. On one hand, PCs make drag-and-drop and touch-screen control possible. They support sophisticated data gathering, transmission, and analysis, as well as integration of 21 CFR Part 11 compliant audit trails.Unfortunately, those strengths are also PCs' weakness. "PCs are not like a PLC's ladder logic," explains R.A. Jones & Co. Inc. sales director Rusty Sparling. "PCs can do 10 different things at the same time, so it's a challenge to validate a machine when all those functions are going on simultaneously."A pharmaceutical machine has to operate the same way every single time. Instead of three paths, there can be only one path to an outcome and it always has to be the same path," Sparling says. "The concern with a PC system is that if one thing goes wrong, the PC will find another path around it."Darren Elliott, R.A. Jones' chief electrical engineer, is even more succinct: "They've seen the blue screen of death on their desktop computer," he says, referring to a frozen PC, "and they're afraid of that occurring in their repeatable process."

A servo-driven C2K cartoner from R.A. Jones. Courtesy of R.A. Jones.

ReliabilityAs a result, some pharmaceutical packagers insist on PLC-operated servo systems. In essence, they dumb down their machinery to keep it from acting up. The PC is still there, but it is segregated and used only for 21 CFR Part 11 security and data logging. The PLCs actually run the machine. This may improve reliability but it also eliminates many of the advantages of moving to servos in the first place.Yet there are other ways to ensure reliability. The key, says Blauw, lies in taking control of the PC processor before Windows can get to it. "When you start a computer," he explains, "the system initializes itself and then releases control of the processor to an operating system, such as Windows. Then Windows boots up and allocates processor time to different applications, such as Word or Outlook."In a computer-controlled packaging line, after initialization is done the control architecture grabs 100% control of the processor before Windows gets there,” Blauw says. It then delegates processor time to Windows. So you can run Windows applications, but it is the control architecture that's in charge. Under this scenario, if Windows dies the packaging line keeps running. You may have to switch to manual control because the [human-machine interface, or HMI] is gone, but you're still running."Even that may seem risky to some manufacturers. They may opt for a combination of PC and PLC controls by using the HMI to switch PLCs on and off. "Many systems use a servo controller that slides into a PC slot," explains Blauw. "The PC software sends commands to a servo controller which controls PLCs. The servos are subservient to the PLC, which is executing the process. The servos then send any feedback on the process back through the PLC."Some companies are attracted to hybrid systems. They like the simplified operation and data collection as well as the repeatability that comes with servos. Separating the PLCs from the HMI also reduces the likelihood of mistakes that affect the validated packaging process.Others don't want to deal with the headaches of PC-based systems. "They just want to know that it will keep running," Blauw admits. "They worry about stability. If they buy a PC to run the front end of the machine, they're cognizant of PC shelf life and want to make sure they can buy a PC part."Still others worry about their ability to support the new technology, and the staffing and training issues that a switch may pose. "The corporate staff may want all-servo machines, but at the plant level they can't support it," says Elliott. "Plants traditionally have a mechanically oriented maintenance staff. It takes an entirely different skill set to support software-drive servos."Blauw likens the transition to moving from carburetor to fuel-injected engines in automobiles: "You used to be able to open the hood and fix it, but that's not an option any more. Skill sets needed for this equipment are completely different. Instead of looking for something that's loose, you might have to diagnose a sensor that's not sensing properly."Standards on the horizonUltimately, the transition to servos may be eased by the push for packaging system standards. Over the past few months, the Open Modular Architecture Controls (OMAC) Packaging Workgroup has teamed with the World Batch Forum (WBF) and Instrumentation Society of America (ISA) to accelerate its vision of plug-and-play packaging modules.OMAC's goal is to set standards that define how packaging line software and hardware communicate with one another. Devices that adhere to these standards, such as servos, would work when attached to any OMAC-compliant packaging line, much the way a graphics card or modem works when attached to a PC.It would be the vendor's job to validate the servo, not the drug maker’s. Once validated, though, drug makers would not have to go through the laborious effort of revalidating each servo. Instead, they would build a system from OMAC-compliant modular components, and then validate the final packaging line.The result would be an open, flexible packaging system that could be reprogrammed to work with a variety of drugs and containers. It could also be rebuilt to perform entirely different tasks.Given the competitive forces pressuring the industry, lower costs and greater flexibility are no longer options. Even validation concerns aren’t likely to slow the move to servos.
About the Author

Alan S. Brown | Contributing Editor