09-07, 15:15–15:55 (Europe/Berlin), Main Stage
Conventional spacecraft cable harnesses are costly, labor-intensive and difficult to integrate, particularly when numerous sensors and actuators must be distributed across a mechanically complex structure.
For a deployable spacecraft system, we replaced much of the conventional harness with a custom rigid-flex PCB designed in KiCad. The spider-like assembly wraps around the mechanical structure and branches out to multiple functional locations, allowing sensors, switches, heaters and interfaces to be integrated without separate point-to-point wiring.
The system was launched into low Earth orbit in March 2026 and has been operating reliably since April, including rigid-flex sections installed on the exterior of the spacecraft structure.
This talk presents the KiCad design workflow, the close integration with mechanical CAD, the manufacturing and integration challenges, and the lessons learned from taking an unconventional rigid-flex design from its first fabrication run to successful operation in orbit.
Cable harnesses are a major cost, integration and reliability driver in spacecraft development. Conventional solutions require numerous individual wires, connectors, crimp contacts, strain-relief provisions and manual assembly steps. Depending on the application, connectors and cable terminations may additionally require securing or encapsulation. Every cable, connector and auxiliary material must also be assessed with regard to its suitability for the space environment, including material compatibility and outgassing requirements.
For a deployable spacecraft structure, we therefore investigated whether a distributed sensor harness could be implemented as a rigid-flex PCB instead. The resulting design consists of a central interconnected structure with multiple flexible branches extending toward individual sensing locations. Sensors such as I²C and 1-Wire temperature sensors are mounted directly at the ends of these branches. The complete assembly wraps around the mechanical structure and provides broad sensor coverage without requiring a conventional point-to-point cable harness.
A rigid-flex implementation can simplify material selection and qualification because the relevant substrate materials, coverlay or solder-mask systems, surface finishes and assembly processes are defined as part of a controlled PCB stack-up. This does not make the PCB automatically space-qualified, but it reduces the number of heterogeneous cable, connector and encapsulation materials that must be selected and assessed individually.
The approach becomes particularly attractive when moving beyond a single prototype toward small- or large-series production. A conventional harness must be assembled repeatedly by hand, including cutting wires to length, stripping, crimping, routing, securing and adding strain relief. These labor-intensive operations must be performed and inspected for every individual unit. Once a rigid-flex design has been validated, however, the same PCB can be manufactured repeatedly from an unchanged design dataset. Producing one hundred units therefore does not require manually reproducing the complete harness one hundred times. This can substantially improve scalability, repeatability and quality consistency while reducing recurring assembly and inspection effort.
A key part of the development process was the close integration of the PCB geometry with the mechanical CAD model. The three-dimensional folded geometry and overall envelope of the flex harness were developed in CATIA. A sheet-metal design workbench, normally used to model bent metal parts with defined bend radii, proved particularly useful for creating the curved, flat geometry of the harness. The resulting three-dimensional shape could then be unfolded into a planar contour and imported into KiCad as the basis for the PCB outline. This workflow made it possible to integrate the flex assembly directly into the mechanical design instead of treating the harness as a separate component added at a later stage. Despite the tight tolerances of the assembled structure, all harness components fitted correctly after the first fabrication run.
The presentation will cover the interaction between CATIA and KiCad, the definition of rigid and flexible regions, routing through curved and mechanically constrained areas, component placement at remote sensor locations, manufacturing constraints, mechanical integration and the design decisions driven by the intended space environment.
The system was launched into a 530 km Sun-synchronous orbit in March 2026 and has been delivering data since April. Parts of the rigid-flex assembly are installed on the exterior of the spacecraft structure and are directly exposed to the space environment, including temperatures ranging from −35 °C to +80 °C and atomic oxygen. The complete sensor network has so far operated without failures.
The project originated from DLR technology-transfer activities and contributed to the formation of ADPERA, a spin-off developing deployable structures for creating large systems in space, such as instrument masts, photovoltaic generators and antennas. It provides a practical example of how KiCad can be used not only to design electronics, but also to replace a mechanically complex spacecraft subsystem with an integrated, flight-proven PCB solution.
Marco studied Mechatronics, giving him a broad interdisciplinary engineering background spanning mechanics, electronics, and software. He began his career at the German Aerospace Center (DLR), where he specialized in deployable spacecraft structures, mechanisms, and lightweight design. His work in this field culminated in a doctoral thesis on a large rollable synthetic-aperture radar antenna for Earth observation and included research stays at NASA and ESA.
Over the course of his career, his focus increasingly shifted toward embedded electronics, firmware, and the integration of complex mechatronic systems. He developed electronics and software for several parabolic-flight campaigns and later for a spacecraft experiment. He has extensive experience integrating mechanical structures, sensors, actuators, avionics, and third-party hardware and software into reliable spaceflight systems.
In 2026, Marco co-founded ADPERA, a DLR spin-off developing deployable structures and integrated systems for creating large structures in space, where he serves as Chief Technology Officer.