Microgrid: Why Nexans’ AmpaCity Experiment Signals a New Electrical Architecture
The experiment conducted by Nexans at its AmpaCity R&D center in Lyon warrants attention beyond the mere product announcement. It highlights a far more structural issue: the potential redesign of electrical architectures for buildings, commercial sites, and—in the future—energy-intensive data centers.
A Shift in Architecture, Not Just in Cabling
Most modern equipment already operates natively on direct current (DC): power electronics, LED lighting, batteries, photovoltaics, charging stations, and computer servers. Yet, building networks remain largely structured around alternating current (AC). This situation necessitates successive AC/DC and DC/AC conversions, which complicate the architecture and generate energy losses.
The AmpaCity microgrid tests a different approach: organizing local distribution around a 700 VDC bus. Developed in partnership with Schneider Electric and VINCI Energies, the demonstrator powers various real-world applications, including workstations, lighting, battery storage, and two 30 kW fast-charging stations. It aligns with the Current/OS interoperability framework, reflecting a commitment to making DC systems compatible across different suppliers and use cases.
Why 700 VDC Is a Stepping Stone to 800 VDC
The technical advantages of direct current extend beyond the elimination of conversion steps. By increasing the distribution voltage, it is possible to reduce the current required for the same power output. This reduction in current lowers Joule losses, eases certain cabling constraints, and paves the way for more compact architectures.
Nexans explicitly envisions a transition to 800 VDC architectures for data centers. This point is crucial: AI workloads are driving up power density per rack, with infrastructure requirements potentially exceeding 100 kW per rack. In this context, power distribution efficiency becomes a critical performance factor, just like cooling or server architecture.
R&D challenges behind the apparent simplicity
Implementing direct current (DC) in buildings presents several challenges. It requires securing electrical protection systems, managing DC arcs, ensuring interoperability, guaranteeing compatibility between power sources and loads, and harmonizing standards. The AmpaCity microgrid is significant precisely because it tests these issues under real-world conditions. The R&D value lies not just in the cable itself, but in system integration: distribution, protection, monitoring, equipment compatibility, and the standardization roadmap.
The prospect of combining 800 VDC distribution with high-temperature superconducting (HTS) cables also opens up new possibilities. HTS cables—capable of transmitting electricity with near-zero resistance under certain conditions—could become viable in environments with extremely high energy density. However, their adoption will depend on costs, cryogenic requirements, operational reliability, and associated business models.
Why this is a key topic for technology monitoring
For GraphMyTech, this development illustrates a classic scenario where monitoring limited to standard keywords risks underestimating the subject's importance. It is not merely a matter of "cables" or "smart buildings." One must map out an entire ecosystem: energy players, equipment manufacturers, integrators, standards bodies, data centers, storage systems, EV charging, and superconducting technologies.
Using specialized monitoring tools makes it possible to detect "weak signals": a rise in publications regarding DC power, patent filings for protection systems, Current/OS pilot projects, alignment with the needs of AI data centers, and the emergence of HTS components. It is this consolidated view that enables innovation departments to distinguish an isolated experiment from an emerging technological trajectory.
GraphMyTech helps structure precisely this type of analysis: state-of-the-art reviews, stakeholder mapping, patent and publication trajectories, identification of weak signals, and support for R&D decision-making.
Comments