Medical Device Link.

Originally Published PMPN June 2003

Heat Sealers

Medical device manufacturers look beyond data acquisition to help control the heat-sealing process.

by Erik Swain

A few years ago, data acqui-sition was all the rage for medical heat sealers. Some medical device manufacturers thought a mechanism to generate continuous data would help improve control of the heat-sealing process. As a result, a number of heat-sealer manufacturers added electronic data acquisition features to their machines.

But the data acquisition trend never really caught on, for a variety of reasons. These days, there is not much demand for those features. “A lot of machine users have come to the realization that the ability to collect information is great, but in the real world, most of it is not needed,” says Charles Trillich, president, Packworld USA Ltd. (Nazareth, PA). “One only needs a limited amount of select data to confirm validation.”

However, there remains interest in improving control of the heat-sealing process and in documenting those improvements. There remains interest in equipment that is easy to validate. And there remains interest in making heat sealing a faster and cleaner process. If continuous data acquisition is not going to be a solution, other things will have to step up in its place.

THE DATA QUANDaRY

It was thought that the ability to collect data on a continuous basis would teach users a lot about the heat-sealing process. The second-by-second output would be a good indicator of the control of the heat-sealing process. And the day-by-day output would be useful in developing a validation plan. 

But in practice, some in industry say, the feature often raised more questions than it answered. It opened users up to more scrutiny from FDA. It took away focus from the engineering of the equipment. And it added cost that many users said was not justified. “I don’t know if we ever sold more than one or two units with those features per year,” says John Lewitt, vice president of sales, Emplex Systems Inc. (Toronto). “Customers can’t justify the cost. It means the validation process is even longer, because you have to stay on top of it to a higher degree than you really need to.”

Kent Hayward, marketing manager for Alloyd Company, Inc. (DeKalb, IL), notes that his company, too, has very few customers that are collecting continuous sealing data. “What users want, in my opinion, is more control over the process and communication from the machine to the operator if the process is out of range,” he says. “To achieve this control, we add a PLC (programmable logic controller) and an HMI (human-machine interface), plus custom software and a light tower, among other things. Keyed entry, password protection, and recipe management have also been popular options for us.”

The core of the issue, says Kent Hevenor, is that “data acquisition does not make the machine a better machine. It only seals as well as it’s built.” Hevenor is the product manager–medical packaging equipment for DT Converting Technologies (Hyannis, MA). Instead, he says, the feature introduces more nonessential variables to the process. Users, he says, do not seem to think those variables add value.

“To get data acquisition, you need a PLC or a microprocessor-based control system,” he says, “and that means software validation. A lot of users do not want to take on that additional responsibility, and they could be responsible for producing that data in an audit. FDA is not telling you to gather it, but if it’s part of your process, they want to know about it. Is that a road you want to go down? A lot have not.”

Continuous data acquisition posed a particular problem for impulse heat sealers, says Liston Bevard, president, Impulse Dynamics Inc. (San Francisco). “Impulse sealing is a dynamic process by its very nature,” he says. “The heat profile along the length and width of the heating element can vary slightly. Residual heat buildup and the methods to control it, such as water-cooling, also contribute to variances in the temperature profile of the element and the heater bar. Monitoring clamping force pressure contributes its own set of values into the process as well. We created a data stream to capture the temperature profile. But then we had to deal with all kinds of variances attributable to the dynamic forces in play. All that data became a headache for people to process in an easy-to-understand manner. They had to make the data understandable to their own organizations and FDA. Users were spending too much time and resources not related to productivity. We had to scale back expectations that these new bells and whistles would easily provide the basis for validating the process. The real interest now is to provide a well-built machine where you can calibrate the instruments and components to what they should be and use this information and seal-strength tests to develop validation guidelines.” 

The PW1036 from Packworld USA (Nazareth, PA) produces validatable seals.

The key to validation, says Stan Hall, Saxon product sales manager, Fischbein-Saxon (Statesville, NC), is documenting that a machine has been set up properly and maintained. “I don’t believe a continuous stream of data collection is important for heat sealing and validating a manufacturer’s product packaging,” he says. “You need to show that you’ve checked the machine, set it up properly, performed regular calibration, produced the product, and documented everything. If you have reams of data, it opens the door for FDA to look at masses of detail and to see if exceptions exist. Any process will occasionally have an exception. That doesn’t mean you produced a bad product. At some point, not only does it not serve the interest of the manufacturers and their customers, but it opens up opportunities for litigation.”

CALIBRATION

So instead, machine suppliers and their customers have begun to rely on simpler calibration methods for time, temperature, and pressure. 

“There are different means to validate the temperature of the heat-seal device as compared with a NIST (National Institute of Standards and Technology; Gaithersburg, MD) standard,” Trillich says. “In a constantly heated device, that’s relatively easy 
to do. For temperature, use a NIST-calibrated thermocouple. For pressure, use strain gauges for mechanical sealers and calibrated pressure gauges for pneumatic ones. For time, use a stopwatch.”

However, he notes, “it is not quite so simple on an impulse seal. The temperature comes up to a predetermined set point, but then it is gone before you have an opportunity to get an accurate reading. Thermocouples do not respond fast enough to capture the peak sealing temperature of the heat-seal band. Our control systems have high-response controllers that use the heat-seal band itself as the measuring device to ensure virtually instant feedback. A permanent record is available using a datalogger to capture the heat-cycle profile. A simple way to validate is to use a NIST-certified calibration strip. It’s inserted between the jaws, and one can gradually increase the set temperature, one cycle at a time, until the calibration strip is actuated by either changing color or partially melting the heat-sensitive material. It’s a very precise means of simulating exactly what the workpiece is exposed to when sealing.”

Testing of tolerances is also an important factor in process control, Lewitt says. “For any of us to say we have a validatable machine, the consistency has to be there,” he says. “You have to do burst and peel tests, to set up the machine on all parameters, and to set tolerances for temperature, pressure, and speed. If you do a test, uniformity is guaranteed by the parameters and tolerances that you choose. If it falls out of those, alarms go off and you can’t seal the bag. Once the bag exits the machine, you know it’s been sealed at the parameters you’ve set it for. The only way it will fail is if the parameters you set it for are wrong.”

Charlie Webb, CPP, validation specialist, Van der Stahl Scientific (Wrightwood, CA), agrees. “A lot of times, people don’t take the whole triad of temperature, pressure, and dwell. But you have to,” he says. “Our new machine has a polarized pressure feedback system, controllable and maintainable, that makes sure pressure can be dealt with as well as temperature. The control attributes are all calibratable. We have large sterile-packaging evaluation labs. We make sure our customers send us their packages and put them through a battery of seal-strength tests. We look for ways to remove nonessential components, but still produce a good validatable machine.”

The goal, says Bevard, is to demonstrate a high level of statistical confidence in the process. The dynamics of the impulse heat-sealing process may mean that the most reliable way to demonstrate that the process is in control is through systematic testing of the seal strength. That means quantifying a control value for seal strength and documenting the test results for each production run. This procedure has historically been the foundation of a validation protocol.

“You may work with an FDA inspector or a QA department that will insist that the machine display the actual temperature reached on the heater element. This is not as easy and straightforward as it may seem when working with the impulse heat process,” he says. “There are those who appreciate the complex characteristics of the process and accept validations based on sound test results and documentation.” 

IMPROVEMENTS

The return to simpler calibration methods indicates less of a need for continuous data acquisition features. As a result, many equipment vendors have turned their efforts to improving the speed and quality of their machines. For example, Packworld has developed a technology that has improved impulse seal times from 0.6 seconds to 0.2 seconds, Trillich says. “It’s a redesign of the heat-sealing system using a new controller,” he says. “It provides shorter ramp-up time to reach the set point more quickly and reduces cooling time by chilling the heat-seal bar. This new system produces a dramatically shortened heat-sealing cycle.” 

On the verification side, Van der Stahl is currently working on a process to determine whether a package is wholly sealed, Webb says. “It’s a visual inspection unit that gives a quick, stable, visual reference if the seal passes visual evaluation,” he says. 

Emplex has come up with a way to seal bags made from less-robust materials just as well as ones made from stronger materials, Lewitt says. “In one instance we developed a new option to seal uncoated bags that had poor seal quality. We tweaked a process in the sealing section,” he says. “The general trend [in medical device packaging] is to save money by going to cheaper bags that still meet your requirements. But you have to figure out how to seal them the same way.”

And the blurring of the drug and medical device industries via products such as drug-coated stents has created new challenges for heat sealers, Hevenor says. “People are asking us for a machine able to evacuate a pouch, or evacuate and gas flush, prior to sealing,” he says. “There are many drug-coated products and more medicated implantables. It’s bridging the requirements of pharmaceuticals and medical devices. Our new machine addresses these issues. That has been the big new thing swirling around the industry over the past year.” 

Copyright ©2003 Pharmaceutical & Medical Packaging News