Quantcast

User Stories

Centre for Concepts in Mechatronics Improves Resolution of Industry-Leading Agfa Printers Using Model-Based Design

View PDF

Centre for Concepts in Mechatronics User Story image
Substrate path of industrial printer.

"It can take lots of time to convert your control architecture to a software implementation because you have to communicate your design to programmers who are not machine dynamic experts. Using MathWorks tools, we can realize our implementations directly, reducing costs and saving up to four weeks of time."

Arend-Jan Beltman, Centre for Concepts in Mechatronics

In a single pass, the Agfa :Dotrix Modular inkjet printer creates more than two-foot wide, full-color images at speeds of more than 24 meters per minute with minimal setup times and job switching. Though already delivering the most productive industrial digital print solution on the market, Agfa approached the Centre for Concepts in Mechatronics (CCM) to improve the printer's repeatability.

Using MathWorks tools for Model-Based Design and its own Smart and Flexible Control Solutions (SAXCS) approach, CCM created a custom controller that increased the repeatability of existing :Dotrix Modular printers.

"We delivered a flexible, hardware-independent solution to Agfa, while reducing development costs," says Arend-Jan Beltman, manager of the Department of Mechatronics at CCM.

Challenge

The :Dotrix Modular system’s print-encoder roll, which carries more than 225 kg rolls of paper, triggers the firing mechanism of ink droplets from a distance of four feet away from the first inkjet head. Tension on the substrate, or printing material, can cause it to stretch, misaligning the ink droplets and reducing precision. In existing :Dotrix Modular printers, operators could not tune the force control properties that are involved in the printing process.

After isolating these variations in substrate forces as the primary factor affecting precision, CCM needed to design, develop, and deploy a control system that could work with existing printers. This would enable operators to rapidly and accurately calibrate the machine’s control parameters for any substrate. At the same time, CCM sought to keep development time and implementation costs to a minimum.

Solution

A team of CCM engineers used Simulink®, Simulink Coder™, and Simulink Real-Time™ to develop, validate, and deploy a sophisticated control system for :Dotrix Modular printers to achieve increased repeatability.

Designing the substrate control system involved tuning three control loops that exhibited coupling. The team first modeled the nonlinear dynamics of the substrate path using Simulink. With this Simulink model, CCM engineers then developed controllers using their own controller design tools supported by Control System Toolbox™. Using Simulink, they simulated and refined the control strategy for the three loops to minimize unwanted loop interaction and achieve the desired printer precision.

After simulating the control system in Simulink, CCM engineers used Simulink Coder to automatically generate C code for the controller. This code was then tested on a real-time prototype using dSPACE® hardware. They then used Signal Processing Toolbox™ to analyze measured data.

Following an iterative development cycle, the team tested the solution on a prototype machine, refined the design in Simulink, and automatically generated code with Simulink Coder for further evaluation and to identify external factors not considered earlier in the process.

The engineers then used Simulink Real-Time to deploy the real-time embedded system on PC hardware for production use in the printer. This enabled them to easily port the validated design to the hardware without translating the code by hand.

Using MATLAB® development tools, the team created a GUI that enables printer operators to easily tune the control system for any new printer substrate. The team used MATLAB Compiler™ to create a standalone version of the GUI that is deployed with the controller.

The CCM control system has been successfully deployed on existing :Dotrix Modular printers.

Results

  • Design iterations completed in minutes. "With Model-Based Design, we see development time reduction, cost savings, and improved time-to-market," reports Beltman. "Automatic code generation saves a great deal of time. In addition, we can simply change the model to update our design, and we know it will be implemented as required. Our design iterations, from making changes to hardware testing, are completed in five minutes."
  • Functional performance improved by a factor of five. "Using MathWorks tools and our SAXCS approach, we improved functional performance by a factor of five,” says Beltman. "The combination of CCM's expertise in the field of machine dynamics and control with sophisticated MathWorks tools allows technology innovation to be brought into practice."
  • Hardware-independent solution implemented. "With Simulink Coder and Simulink Real-Time, you can target any PC," says Beltman. "That kind of hardware-independence enables purchasing departments to negotiate on price because they don't have to redesign the system for different controller hardware."

Challenge

Improve image resolution on industrial inkjet printers

Solution

Use MathWorks tools for Model-Based Design to design and automatically generate code for a new control system that enables increased image resolution

Results

  • Design iterations completed in minutes
  • Functional performance improved by a factor of five
  • Hardware-independent solution implemented

Products Used

Learn more about Centre for Concepts in Mechatronics