After 34 months of research, the 3D-FiberTrain consortium has demonstrated a new approach to manufacturing large, highly loaded composite components for rail vehicles. Led by Hörmann Vehicle Engineering (HVE), the project ran from September 2023 to June 2026 and targeted an alternative to conventional thermoset composite processes such as hand lay-up and vacuum infusion.



The process chain combines three manufacturing stages. Large-format, granule-fed extrusion printing first produces the complex thermoplastic geometry without dedicated mould tooling. A newly developed automated 3D tape-laying process then adds continuous-fibre reinforcement selectively in highly stressed areas, following the component’s load paths. Automated milling completes the process to achieve the required dimensional accuracy and surface quality.


The entire manufacturing chain is thermoplastic-based and designed to enable material recycling. For the demonstrators, the consortium used a glass fibre-reinforced polycarbonate specifically selected and modified to meet the stringent fire-safety requirements of railway applications. Fraunhofer researchers also used process simulations to predict thermally induced distortion and delamination during printing.


The technology was validated using two components from Siemens Mobility’s Velaro MultiSystem, operated by Deutsche Bahn as the ICE 3neo: a front skirt and a nose section. The train has an operating speed of up to 320 km/h.



By eliminating dedicated moulds and combining additive manufacturing with targeted continuous-fibre reinforcement, 3D-FiberTrain is intended to reduce manufacturing costs and lead times while allowing designs to be adapted more easily. HVE sees particular potential for small and medium production runs, where the cost of conventional composite tooling can represent a significant constraint.


Alongside HVE, the project involved Fraunhofer IWU, Fraunhofer IMWS and Lakowa, with Siemens Mobility participating as an associated partner.

Photos: Hörmann Vehicle Engineering