
Automated quality assurance in 3D printing
According to Simina Fulga-Beising, scientist at Fraunhofer Institute for Production Technology and Automation (Fraunhofer IPA, Stuttgart, Germany), great hopes the Industry 4.0 called process of intelligent productsion are pinned to 3D-printed components. “They combine functionality, flexibility, complexity and individuality,” she says. They enable the production of individualized workpieces with complex geometries and integrated functionalities, including prostheses or orthoses that fit the wearer’s body perfectly. In addition, 3D printing is the only production process that is already completely digitally controlled. Lot size 1, which industry strives for 4.0, can be realized with it, confirms Fulga-Beising.
So far however, additive production has one catch: quality. Fulga-Beising criticizes the fact that “there are no established standards for overall quality assurance”. Therefore, safety and reproducibility cannot be guaranteed – though in industries such as medical technology, such specifications are extremely important. In addition, the lack of quality control during printing also causes high costs for the company. The printer works completely self-sufficiently. In the worst case, users will only notice an error or failure after the component is finished. Having the process monitored by a technician would be far too expensive in view of the long running time of the machine.
With IQ4AP, the Fraunhofer IPA has developed a system that automatically controls the quality of 3D printing inline – i. e. during printing. The application is based on a “black box” containing a camera, lighting and ventilation. Key technology is machine vision. A camera system scans the freshly applied powder layers and sintered layers directly in the process. The images are then checked with several algorithms. “Coarse and fine defects are detected immediately. Even the characteristics of the sintered layer, such as lengths or hole diameters, can be measured inline. This produces a component quality protocol at the layer level,” said the scientist. The machine operator is automatically notified by SMS or e-mail, for example, and can decide what to do. Tolerances, for example the maximum admitted distance between holes, can also be defined. The process can now be validated. “With industrial computed tomography, we were able to confirm the results of the inline quality control system,” said Fulga-Beising.
The prototype for the inline quality control system with adaptable process-integrated sensor measuring technology using selective laser sintering (SLS) as an example, was realized by IPA scientists in 2016. The hardware costs users only 2500 €. It is machine-independent and can be docked to any 3D printer. No cooling is necessary to protect the hardware components against the high temperatures in the pressure range. The system can be used immediately, without time-consuming and costly machine certifications.
Theoretically, the module can also be adapted for quality control in the metal sector. The IPA researcher has developed a corresponding software and hardware concept. IQ4AP is also modular and can be extended.
The IPA is now looking for partners who want to test the system and integrate it into joint projects according to their needs. The researcher is pleased to announce that “initial enquiries have already been received.
However, from the researcher’s perspective, the work on the inline quality control system is far from complete. In the next step, the system will apply machine learning to assess for itself which consequences an error will have for the printing process. This includes not only deciding whether to stop it, but also drawing conclusions and optimising the procedure.
More information: www.ipa.fraunhofer.de
