Videos and discussions on how sensors and software work to improve composite productivity #trends #automation #layup

Automatic fiber placement (AFP) expert Coriolis Composites (Quéven, France) demonstrated an online inspection developed in collaboration with Edixia (Vern-sur-Seiche, France), which provides automatic inspection and industrial vision technology. The sensor manufactured by Edixia is integrated into the placement head of the Coriolis composite AFP machine. "The solution can detect all standard defects, including gaps, overlaps, twisted tows and pompoms," said Godefroy Marechal, Coriolis Composites R&D Project Manager. "In the continuous production process of complex-shaped parts, it should allow 100% inspection and increase productivity by 20-30% compared to current methods. The system is easy to configure and can be used for a variety of composite materials, including carbon fiber Reinforce thermoset materials, thermoplastics and even ceramic materials."

Marechal explained that Coriolis Composites has wanted to implement online inspections for some time. "We understand that our customers need this solution, and all AFPs must be integrated for the future," he said. "Three to four years ago, when Coriolis started using automatic detection, we thought we would find an off-the-shelf plug-and-play system, but then realized that it didn’t exist yet. Edixia is a good partner, Because it has more than 35 years of experience in vision systems and surface inspection. A few years ago, we worked with Edixia to use one of their vision systems in the automatic calibration of our AFP robots. So now we have combined the two The company’s expertise has developed an efficient but powerful solution."

The sensor is based on Edixia's AF-Inspect technology and includes a line laser and a line-shaped camera. "It's like a profiler," Marechal said. (For more information on profilers, please refer to "Automated in-situ inspection is a necessary condition for the next generation of aerospace"). "From that shape, it will generate a data point cloud."

The image above shows the housing (housing) of the Edixia sensor and its purple laser line, which is used to scan the surface of the AFP laminate for defects. The external encoder connected to the AFP head, the sensor knows its position in the 6-axis 3D space. Source | Coriolis Composites

But before analyzing this data, the point cloud must first be directed to the 3D reference system of the Coriolis composite AFP head. "The Edixia sensor is connected to the external encoder of our head, and its 6D position (x, y, z, A, B, C) is obtained in real time through our algorithm." This is similar to the programming method of CNC machine tools, with 6 free axes Is the X and Y axis in the plane, and the Z axis direction out of the plane. A, B, and C are the degrees of rotation around each of these axes. The data is then analyzed through a combination of Edixia and Coriolis Composites software, which identifies defects in real time based on a combination of algorithms.

The Edixia sensor is easily attached to the Coriolis Composites AFP head without removing it from its robotic arm. Source | Coriolis Composites

"Adding a sensor is physically easy," Marechal said. "You only need to connect it with fasteners, you don't need to remove the AFP head to add it." Communication in software is the challenge and will continue to evolve . "We are working hard to make this solution seamlessly connect with our customers," he added.

Coriolis Composites also collaborated with Edixia to integrate online detection sensors with minimal changes in the shape and volume of the AFP head. “The Coriolis composite AFP head is compact, which helps customers not only want to produce flat panels, but also want to produce complex 3D shapes,” Marechal said. "This small head helps avoid collisions with flanges and corners. For parts such as spars, this ensures the quality of fiber placement in the corner area."

However, inspections in these areas are a key issue. "There are very small areas in the corners, and you cannot check these areas at the moment," Marechal points out. "We have considered how to use the Edixia sensor in combination with an additional, inspection-only robot path in order to be able to detect 100% of such parts. It is difficult to obtain the correct sensor and the correct algorithm, and the correct system as a whole To make complex 3D parts. But that’s what we provide. This is what I’m trying to express in the video: we can perform 2D inspections on flat plates and 3D spars and parts with complex shapes and corners. Inspection. Customers can come to our R&D facility and we can test specific parts to prove that we can provide 100% inspection."

Once the sensor and AFP head are capable of 100% inspection of the part, the next step is to identify defects. "It's the job of the software to retrieve the data and determine what the defects are, where they are located, and their impact on the entire part. This is what we are refining now."

In the screenshot below, you can see the Edixia sensor output on the left. "This type of image will be displayed after each tape is laid," Marechal said. "Pink indicates the gap. For example, the long pink line on the left indicates the gap between two tapes. The blue box shows the defect in the Z direction. In the tow, you can see twisted and non-straight fibers."

The Coriolis composite human-machine interface (HMI) is shown on the right. "The Edixia sensor is integrated into our AFP machine and then communicates with our HMI, allowing the operator to indicate the repair location of the part in real time," Marechal explained. "This is what we are doing this year to improve the communication between Edixia sensors and our software. We have developed an interface between the two and are now conducting trials to verify this and enable the software to recognize and annotate each Defective algorithm. We have established a partnership with our customers to complete this verification in the next few months."

Screenshot of Edixia sensor software (left) and Coriolis Composites human-machine interface (HMI, right). Source | Coriolis Composites

What is the ability of the sensor to show the location of the defect? "This is not a problem," Marechal said. "We have developed the ability to communicate with laser projectors and cooperated with LPT (Laser Projection Technologies, Londonderry, NH) in the United States, but then they were acquired by FARO Technologies (Lake Mary, Florida, U.S.), which changed everything about their software. . So, now I will have to redevelop it, but it’s very simple."

He continued, "However, we have tried to cooperate with other laser projection systems, but we are not limited to one supplier. It is just a question of performance and price and what our customers want." How to integrate with virtual reality (VR) To achieve 3D visualization of defects? "At present, this is not what our customers emphasize, but if needed, we can help them provide them with this ability in the future."

The path of the Edixia sensor can be seen as a purple laser line (left), in the simulation (right) it can be seen as a purple trapezoid. Source | Coriolis Composites

Another focus of Coriolis Composites is the ability to use these defect data to optimize the AFP process in its internal digital chain. "This is what we want to tell our customers: We have developed our own internal software and fully managed our own digital process chain, fully predicting the continued development of the inspection strategy," Marechal explained. "This is part of what I was referring to when discussing our offline programming software in the video."

I discussed this in my 2018 feature article "Integrating Thermoplastic Aerospace Structures in Place, Part 1":

Coriolis Composites is integrating online inspection sensors in its entire digital process chain, including simulation and optimization of the layup process. Source | Coriolis Composites

"...Parts produced on Coriolis machines now benefit from what the company calls a "closed simulation chain", which integrates computer-aided design, manufacturing and engineering (CAD-CAM-CAE), and is interfaced by the company’s two-way software And integration support. “Parts started with OEM design in CATIA [Dassault Systèmes],” (Coriolis CTO Alexandre) Hamlyn said. “Our CAT/CADFiber software imports composite stacks from CATIA and provides users with modeling of all fiber structures. Tools needed. Then it generates and optimizes the AFP tape/trailer route. " 

After producing the test parts for verification, the software will export the true "as-built" fiber angles to commercial FEA solvers (such as NASTRAN, ABAQUS, SAMCEF) and enable the mesh between the AFP surface and the structural FEM mesh Mapping. "

"We use standard Siemens controllers," Marechal admits, "but our software is the key to using AFP machines with complex 3D parts. Our control of the digital process chain makes the production of composite parts more efficient by ensuring traceability. Reliable and now further optimized. For example, if defects continue to occur, the software will show this and help redesign parts and/or manufacturing to alleviate the situation."

Currently, the system can provide non-contact detection and location of defects larger than 0.5 mm. However, Marechal points out, "Every customer has different tolerances, so no one definition applies to everyone. Every customer needs to customize this. In addition, customers will have to change the tolerances of different parts." It also pointed out that online inspections will find more defects than manual inspectors, "so some adjustments must be made to the current tolerances to maintain production yields. This is also something we are working with customers."

Although Edixia lists a maximum detection speed of up to 2 meters per second, Marechal said it has not yet been tested. So, is it possible to run at full speed with AFP and still collect 100% defect data? "Yes," Marechal said, "but the AFP speed depends on the material you use. Some materials are more difficult to place, so you run slower. The inspection speed also depends on the resolution you need. Currently, we can use 1 meter We have asked Edixia to enable the sensor to run at a speed of 1.5 m/s. We have the idea to further improve accuracy and speed.” Coriolis Composites’ goal is to equip commercial machines with online detection by 2021.

Magnetic microwires are capable of non-contact measurement of temperature and pressure during curing and use.

An initiative to recycle wind turbine blades includes the use of recycled glass fiber composite materials to make cement, substitute raw materials, and save energy.

Stratasys has proven that carbon fiber-wrapped print cores can be used for tool-less, customizable manufacturing, aerodynamic prototypes and production parts for winning race cars.

© 2021 Gardner Business Media, Inc. Privacy Policy [Login]