Solar panels are usually a one-layer-per-footprint proposition: more power means more ground. An Italian startup called A.M.P.S., working with researchers from the “Mediterranea” University of Reggio Calabria, built a prototype that challenges that assumption directly — stacking two layers of bifacial solar mini-modules into the same ground footprint a single layer would normally occupy, with the top layer independently sun-tracking while the bottom stays fixed. The entire motion-and-sensing system is managed by a single Arduino GIGA R1 WiFi board.
The hardware doing the actual work
The GIGA R1 WiFi sits on a custom-designed control PCB acting as the prototype’s central embedded manager, coordinating twelve NEMA23 stepper motors (through dedicated DM542T drivers) that individually adjust twelve bifacial mini-modules spread across both layers, while twelve INA260 sensors — one per module — feed back real-time current readings that the board uses to decide how to orient the tracking layer. Those synchronized current sweeps get uploaded to a remote server for analysis, which is how the team arrived at their headline numbers.
What the stacking actually buys you
Co-founder Cosimo Borrello put the board choice simply: “We chose the GIGA R1 WiFi because it brings together everything a prototype like ours needs, with lots of I/O, a fast dual-core microcontroller and built-in Wi-Fi on a single board” — a fairly ordinary maker-hardware justification for what’s actually a genuine renewable-energy research result. Under clear-sky field testing, measuring short-circuit current (which the team is careful to note isn’t the same thing as total energy production), the double-layer setup produced:
- About 42% more output than a standard single-layer monofacial installation on the same ground area.
- About 27% more than a single-layer bifacial installation on the same footprint.
- A simplified open-loop tracking model, based on a five-parameter sigmoid function rather than full closed-loop control, got close behind at roughly 39% and 25% respectively — suggesting a cheaper, simpler control scheme could capture most of the gain without needing the full sensing-and-feedback loop.
The team is explicit that rooftop, utility-scale, and agrivoltaic applications weren’t evaluated in this field test — this is a proof-of-concept measurement, not a deployed product — but the underlying result (meaningfully more energy density from the same land, measured rather than modeled) is exactly the kind of thing that matters as solar siting runs up against land-use limits. The work is also peer-reviewed, not just a company press release: it’s published in the journal Renewable Energy.
Why this belongs in the “creative hardware” lane
It’s easy to read “two-layer solar array” as purely an energy story, but the build itself is a clean example of what Arduino-class hardware is actually good for in a research context: a dozen independently-controlled stepper motors, a dozen simultaneous current sensors, and a wireless uplink, all coordinated from one board that a small team could design a custom PCB around rather than engineering a bespoke embedded system from scratch. That’s the same pattern that shows up across kinetic sculpture, robotics, and interactive installation work — just pointed at a genuinely practical research question instead of an art piece.