Samenvatting

One of the recent developments in the aerospace industry is the use of thermoplastic composites. Thermoplastic composites can be welded instead of using other bonding techniques. One of the welding techniques is induction welding. Induction welded parts are already used in aerospace applications, yet the technique is quite new and not completely developed. No research has been done or is archived in induction welding of parts with variable thicknesses. Most of the welding parameters are found by “trial-and-error” and have a constant thickness.
The main goal of this project is to gather knowledge and data of welding composites with variable thicknesses. After gathering all the data, a weld can be made with a variable thickness.
The main question is specified as: What is the best way to bind thermoplastic composites by induction welding?
To get an answer to the main question, desk research has been performed to gather information about thermoplastic composites and induction welding. With the information gathered in the desk research, some sub questions ca already be answered. With the information gathered during desk research, test plans are made to get answers to the other sub questions;
Welding composites with variable thickness would require variable parameters. Tests have been performed to analyze the effect of welding speed. This test shows that different speeds can be used for induction welding of thermoplastic composites. Welds are made with three different speeds, the welds are analyzed with C-scan, microscopic research and with tensile tests.
The influence of thickness is analyzed by heating up stacked plates with thermocouples between them. This shows that the desired weld temperatures can be reached in most depths of the weld area. When this test was performed at different speeds, it was shown that different speeds would give a different depth of where the highest temperature will be reached. Weld interfaces that are located too far away from the induction coil cannot be welded with the setup that was used.
By heating up thicker laminates and combining them with thinner laminates, data from the previous test is validated. If the data from the welds are compared it shows that the temperature that will be reached is dependent on the total thickness or mass of the weld area.
After the basic tests have been performed, a test with milled plates is performed. The plates that are used in this test are milled to create scarf like steps. The scarf like steps are used to make weld interfaces at variable depths. This test shows that some edge effect is present at the scarf-like locations but without overheating material around the weld zone. After testing, the plates are welded using data from previous findings. Some parts of the weld zoned required lower speeds to reach the desired process temperature. C-scans show a good result, microscopic photo’s that were taken at an intersection parallel to the weld direction show that in the middle of the weld line, almost no voids are present and the weld interface is barely visible.
The last test performed is about welding plates with multiple scarfed like steps. For this weld, previous findings are used and this test required variable speed and current settings. The used settings are found by “trial and error” and cannot be mathematically predicted yet at NLR. C-scans are performed which shows good results.
The main conclusion of this report is that parts with variable thickness are weldable by using variable speed and current settings. These settings are highly dependent of the depth of the interface relative to the induction coil. To weld parts with variable thickness, the right speed and current has to be found by trial and error and they can be derived from the tests performed in this research.