KUKA and Magdeburg-Stendal University of Applied Sciences have commissioned an industrial robotic friction stir welding platform intended to let researchers develop joining and machining processes under conditions closer to factory production than a conventional laboratory system.
The research cell developed with KUKA Advanced Welding Solutions is centred on the KR FORTEC ultra MT robot, a heavy-duty platform developed for applications requiring high process forces, rigidity, and precision. The installation combines friction stir welding and milling within one robotic environment.
Friction stir welding differs from conventional fusion welding because a rotating tool generates frictional heat and mechanically works material together without fully melting the parent metals. The process is widely associated with aluminium and other applications where manufacturers require high-strength joints with comparatively low distortion.
Those characteristics have created applications across transport, battery systems, aerospace structures, and other lightweight assemblies. The process also places substantial mechanical loads on its equipment because the tool has to maintain controlled force while following the joint accurately.
Robot stiffness, positional accuracy, force control, component restraint, and tool behaviour therefore become part of weld quality. A system that performs adequately during low-load handling may not maintain the same path accuracy when subjected to sustained friction stir welding forces.
The Magdeburg installation uses production-class hardware rather than a reduced laboratory rig. Researchers can investigate process windows using equipment whose loads, controls, motion, and mechanical behaviour are more representative of systems that could later be installed in industrial production.
Prof Dr-Ing Frank Trommer, Director of the Institute of Mechanical Engineering at Magdeburg-Stendal University of Applied Sciences, said: “With the KR FORTEC ultra MT, we are bringing industrial FSW technology directly into research.”
The KR FORTEC ultra MT also supports machining, allowing welding and milling processes to be developed in the same cell. A component can therefore move through a broader sequence without being transferred to separate equipment simply because the research project has crossed from joining into finishing.
Combining the processes exposes interactions that can disappear when each operation is examined separately. A weld can meet its own quality target while still creating distortion, geometry, or finishing requirements that complicate the following machining stage.
Industrial production eventually has to absorb both operations. Research that accounts for the complete sequence can identify whether a joining process improves the finished component or merely transfers cost and difficulty further along the route.
The cell also includes continuous process data capture and a digital twin. Engineers can use the virtual representation to plan and optimise processes before transferring changes to the physical system, while measured data from the robot can be used to compare the model with actual behaviour.
The accuracy of that connection determines whether the digital twin is useful. Friction stir welding involves interacting variables including robot position, axial force, travel speed, tool rotation, component restraint, material condition, and thermal behaviour. A model that does not reflect those relationships closely enough can optimise an ideal process that the physical cell cannot reproduce.
Consistent data capture gives researchers a basis for examining process robustness as well as whether a successful weld can be produced once. Tool wear, parameter drift, dimensional variation, and component geometry can all be studied over repeated runs.
The platform will also be used for teaching in mechanical engineering, mechatronics, and industrial engineering programmes. Students gain access to an industrial robot and manufacturing process architecture rather than training equipment designed primarily to demonstrate automation concepts at low load.
Companies evaluating friction stir welding face a similar need for realistic conditions. Joining trials can produce convincing laboratory samples while leaving unanswered questions around cycle time, component size, fixturing, finishing, inspection, and integration with existing factory equipment.
A shared industrial research platform allows more of those questions to be answered before a manufacturer commits capital to a dedicated production cell. Parameters, tooling, robot paths, component design, and subsequent machining can be developed together rather than discovered sequentially after equipment has been purchased.
The arrangement also gives designers a route to consider joining constraints earlier in component development. Lightweight structures achieve less benefit when an efficient material choice creates joints that are expensive, difficult to inspect, or unsuitable for automated manufacture.
KUKA describes the new robot generation as suitable for the high forces associated with friction stir welding and machining, with digital tools intended to shorten transfer from development into industrial application. The Magdeburg platform will provide a practical test of that claim across research projects rather than one predefined production part.
Manufacturing research frequently becomes difficult at the point where a technically successful process has to become repeatable, maintainable, and economical. Magdeburg-Stendal has placed those production conditions inside the research cell rather than postponing them until a company attempts to industrialise the result.
