Five hundred meters below the Pacific, off the coast of Los Angeles, a hollow concrete sphere the size of a small house will soon sit anchored to the seabed. When the grid needs power, a valve opens, seawater rushes in, and a turbine spins. When surplus electricity needs somewhere to go, the water is pumped back out. No lithium. No rare earth metals. Just concrete, pressure, and physics. This is StEnSea (Stored Energy in the Sea), a project from Germany’s Fraunhofer Institute for Energy Economics and Energy System Technology, now heading for its first full-scale ocean test off Long Beach, California, by the end of 2026.
The scale of what researchers believe is possible makes the concept hard to dismiss. Fraunhofer estimates that deploying this technology at suitable coastal sites worldwide could unlock a global energy storage capacity of 817,000 gigawatt-hours. For context, Germany’s entire fleet of land-based pumped-hydro plants holds less than 40 gigawatt-hours combined. The ten best European sites alone, according to the institute’s own geographic analysis, could deliver 166,000 gigawatt-hours.
The Long Beach test follows a proof-of-concept trial with a three-meter sphere in Lake Constance, on the German-Austrian-Swiss border. That smaller experiment confirmed the core mechanism works. What remains unproven is whether it holds up under true offshore conditions, at depth, at scale, and over time. That is what this deployment is built to answer.
How a Hollow Concrete Sphere Stores Electricity
The physics come from pumped-hydro storage, a technology grid operators have relied on for decades. Conventional plants pump water uphill when electricity is cheap, then release it through turbines when demand spikes. StEnSea applies the same logic underwater, replacing the mountain with ocean pressure.
An empty sphere on the seabed is a charged unit. Open the valve, and seawater floods in through a pipe under roughly 60 atmospheres of pressure, the weight of 600 meters of ocean bearing down from above. That force drives a pump-turbine in reverse, spinning a generator whose output travels by cable to the shore grid or to a nearby offshore wind platform. To recharge, the pump-turbine switches direction and pushes the water back out against that same pressure. The cycle can run continuously.
Efficiency across a full charge-and-discharge cycle runs between 75 and 80 percent. That is slightly below conventional pumped-hydro, but it sits comfortably within the range of long-duration energy storage technologies currently competing for grid contracts.
Why the Ocean Floor Solves a Problem Land Cannot
Dr. Bernhard Ernst, Senior Project Manager at Fraunhofer IEE, put the problem plainly: “Pumped-hydro plants are particularly well-suited for storing electricity over periods of several hours to a few days. However, their expansion potential is severely limited worldwide. We are therefore transferring their operating principle to the ocean floor, where the natural and ecological restrictions are far lower.”
The institute’s engineers settled on 600 to 800 meters as the target depth range, and the reasoning is practical. Pressure at those depths is strong enough to make storage worthwhile. Standard submersible pumps work reliably there. And ordinary structural concrete, rather than any specialized deep-sea formulation, is sufficient for the sphere walls. A GIS-based coastal survey identified viable sites off Norway, Portugal, Brazil, Japan, and both U.S. coasts. Flooded open-pit mines and deep natural lakes could also host the technology, extending its reach well inland.
The California sphere will be built using 3D concrete printing by Sperra, a U.S. startup focused on additive construction for renewable energy projects. The pump-turbine at the core of each unit will come from Pleuger Industries, a German-founded firm based in Miami and a leading manufacturer of deep-water submersible pumps.
The Numbers Behind the Business Case
Fraunhofer’s cost estimates are modeled on a reference storage park: six spheres, 30 megawatts of combined output, 120 megawatt-hours of total capacity, running roughly 520 charge cycles per year. At that scale, the institute projects storage costs of around 4.6 euro cents per kilowatt-hour, with capital costs of 1,354 euros per kilowatt of power and 158 euros per kilowatt-hour of capacity.
A concrete sphere is designed to last 50 to 60 years. The pump-turbine and generator, which take the wear of each cycle, would need replacing every 20 years. Fraunhofer points to two ways operators could earn a return: energy arbitrage, buying power cheap and selling it dear, and frequency regulation, where grid operators pay storage providers to help keep supply and demand in balance.
What the California Deployment Has to Prove
The Long Beach project is not a product launch. It is a structured test of the full process, covering fabrication, installation, operation at depth, and maintenance, with a specific question at its center: can a nine-meter solution be engineered up to 30 meters without losing what makes it work?
A 30-meter sphere would store far more energy per unit. A commercial park at that scale would be a serious grid-scale storage asset. According to Interesting Engineering, Fraunhofer’s projections suggest a fully built-out system could supply enough electricity to power tens of millions of homes across Europe for a full year, though that figure reflects theoretical maximum potential across optimal global sites rather than near-term deployment plans.
“With the StEnSea sphere storage system, we have developed a cost-effective technology that is particularly well-suited for short to medium-term storage,” Ernst said. “With the test run off the U.S. coast, we are taking a major step toward scaling and commercializing this storage concept.”
Enjoyed this article? Subscribe to our free newsletter for engaging stories, exclusive content, and the latest news.