Simulation software boosts machine productivity
Application Update: Armor increased machine productivity 20% with simulation software to create improvements and decrease downtime.
Armor increases machine productivity 20% by using simulation software to create improvements, rather than trial and error, which can waste time and product.
With sales of 137 million Euros in 2011, Armor produces inked ribbon used in thermal transfer printing for product identification and other applications. Ribbon is produced by applying ink to polyethylene terephthalate (PET) film in a web coating process in which the speed, tension, and position of the web and other variables must be closely controlled to ensure the highest possible quality while maximizing throughput.
Gildas Hubert, project manager for Armor, has simulated many of the company’s coating machines and control systems using multibody dynamics software and multi-domain modeling and simulation software.
“Simulation helped us work out the optimal coating conditions and make engineering changes to our machines,” Hubert said. “Over one year we improved productivity by 20% while also increasing quality of the finished film. Simulation is a great way to improve our manufacturing process at a relatively low cost without disrupting production as is required for physical experiments.”
Based in Nantes, France, Armor was one of the first companies to manufacture carbon film, introduce ribbon cassettes for typewriters, and introduce thermal transfer technology in the early 1980s. The company has more than 760 employees worldwide and produces 110,000 thermal transfer film rolls per day at five production sites around the world. Armor is the leading producer in Europe with a 53% market share. The company offers more than 12,000 ribbon configurations.
Thermal transfer technology
Thermal transfer printing consists of applying thermofusible ink using a heat source emitted by the printer. The thermal transfer ribbon passes over the thermal printhead with the coated side pressed against the label surface. The heat energy produced by each dot causes the pigment to transfer off the carrier film and bond to the surface of the label. The largest application by far for thermal transfer printing is the marking of individual products during manufacturing with information including model number, serial number, use-by date, composition, price, etc. Other applications include flexible packaging, ticketing, personal identification, and plain paper fax machines.
During the manufacturing process, a transparent PET film is unwound as a single or several layers of ink are applied on one side and a protective layer called the backcoating is applied on the other side. The PET film used as the carrier has a thickness of 3.2 mm to 5.0 mm, high resistance to tearing, good thermal conductivity, and very good heat resistance. The backcoating protects the printhead as the ribbon unwinds, provides high thermal conductivity to transfer heat to the print medium, and reduces the formation of static electricity. Different inks are used, including wax, wax-resin, and resin types.
A rubber-coated metering roll feeds the ink to a gravure roll, which in turn feeds the ink to a format transfer roll onto the web. The coating weight is controlled by the velocity of the rolls and the footprint between the metering roll and the gravure roll. All rolls are heated with thermo oil. A jumbo roll 20 km long is coated and then the jumbo roll is unwound onto smaller rolls as required for customer applications.
Moving from physical experiments to simulation
“We have always been concerned with eliminating defects to ensure a positive experience to our customers while at the same time increasing the productivity of our web coating process,” Hubert said. “In the past the primary method of improving operations was with physical experiments. But there were several problems with this approach. First of all, utilizing coating machines to run physical experiments disrupts our production operations. The limited time available for and high cost of physical experiments greatly reduces the number of different conditions that we can evaluate. Physical experiments also provide only a very limited amount of diagnostic information. The number of physical measurements that can be captured during these experiments is limited by the difficulty of instrumenting the coating machines.”
Armor has long been interested in using simulation to evaluate a much larger number of process conditions while reducing the need to disrupt production operations. But in the past the company found it difficult to model the complicated mechanisms and motion control systems involved in roll coating. This challenge was overcome with simulation software that enables control systems to be integrated into mechanical system simulations to optimize system performance.
The multibody dynamics simulation software used automatically formulates and solves the equations of motion for kinematic, static, quasi-static, and dynamic simulations. A graphics-based software tool modeled multi-domain dynamics systems characterized by differential, difference, and algebraic equations used in digital and analog control systems. Integration is accomplished with an interface block in the software model that provides inputs between the software packages.
Optimizing roll coating performance
Hubert constructed a software model of the machine. He defined the rolls as cylinders and added connections between them to represent the gearing in the machine. He defined the material properties of the PET web and entered the friction between the web and the rolls based on physical measurements. The software also simulates the proportional–integral–derivative (PID) closed-loop motion controller.
He found it “very easy to define both the physical and control model” using the software. Hubert began his simulation efforts on a machine whose performance he felt left considerable room for improvement. The machine required continual adjustments to avoid defects. He began by simulating the machine’s current operating conditions. Comparing the simulation results with physical measurements, particularly of web tension, showed that the simulation accurately represented the machine performance.
The simulation results showed that a small change in operating conditions could cause the machine to produce defects. Hubert evaluated changing the operating conditions, particularly the PID control values. He modified the model and re-ran the simulation multiple times, seeking to move the machine to a point where small changes in operating conditions would have no impact on quality. In the end, he discovered more robust operating conditions that substantially improved throughput of the machine by reducing downtime required for adjusting operating conditions.
Based on this success, Hubert turned his attention to other machines that were seemingly operating well to see if improvements could be made in either throughput or quality. During this process, he discovered the importance of accurately determining the friction between the web and the rolls to provide accurate simulation results.
He evaluated more of the company’s machines to identify optimal operating conditions. He evaluated different products with varying film thicknesses on each machine. For each product he evaluated different PID control values to identify values that provided stable operating conditions without defects. During this process, he optimized the control values for each film thickness. This required far more simulation runs than would have been possible with physical experiments. Running virtual experiments with the software also eliminated the cost of downtime on production machines.
Precise PID values
By optimizing control values, Armor increased throughput of its coating machines collectively by about 20% over a one-year period. The primary improvement came from increasing machine reliability and stability so that less time was required for repairs or adjustment. Web speed improvements were also achieved on many machines.
“We are now able to set the PID values much more precisely to optimize the performance of the machine for specific products,” Hubert said. “We have made other improvements with simulation such as increasing the throughput of a cutting machine by 8%. We have plans to apply simulation to additional processes such as our rewinding machines. We are looking to improve the precision of our models by integrating process-related thermal phenomena into the analysis loop. Finally, we also see the potential for simulation to improve the quality of our labeling machines.”
“Our goal was to improve the coating process, notably by controlling the tension of the film onto which the ink is deposited,” Hubert said. Coupling the multibody simulation software with multi-domain modeling and simulation software turned out to be the “ideal way to model our roll machines and control systems. By simulating the operation of our machines, we were able to determine the ideal parameters for operating them over a broad range of products. These calculations make it possible to run each machine at the optimal coating conditions. The end result was that we improved quality while at the same time made substantial improvements in productivity.”
- Chris Baker is the product manager for Adams and Easy5 at MSC Software Corp. Prior to joining MSC Software Corp., Chris Baker worked at UK-based Romax Technology, opening the company’s first American office in Troy, Mich., and leading its business development activities for the Americas. Before joining Romax, Baker worked at Ford Motor Company and spent several years in the Vehicle Dynamics organization. A licensed Professional Engineer and Six Sigma Green Belt, Baker holds a BSE in Mechanical and Aerospace Engineering from Princeton University and an MSE in Mechanical Engineering from the University of Michigan, Ann Arbor. Edited by Mark T. Hoske, content manager, CFE Media, Control Engineering, email@example.com.
See related article below: Simulation and modeling software
- Events & Awards
- Magazine Archives
- Digital Reports
- Global SI Database
- Oil & Gas Engineering
- Survey Prize Winners
- CFE Edu