A 3.1 MWh grid battery energized in eastern Wisconsin runs with no cooling system, and its passive sodium ion design outperformed lithium iron phosphate on a 15-state grid
In a flat corner of eastern Wisconsin, a steel container the length of a semi trailer was switched on and began pushing electricity onto a live grid.
It stored 3.1 megawatt hours of energy.
It weighed 100,000 pounds fully loaded.
A prior installation at a solar testing facility in Watkins, Colorado stores 3.5 MWh and has been running since 2025, but Watkins was a controlled test environment.
And it had no cooling system at all.
Every lithium iron phosphate battery bolted to this same grid needs one, and cooling is where most of those batteries eventually fail.
So what happens when grid storage is built to run without it?
The cooling problem hiding inside every lithium iron phosphate battery storage site
Lithium iron phosphate cells generate heat when they charge and discharge, and that heat is the enemy of both safety and longevity. Left unmanaged, it accelerates capacity loss and, in the worst cases, triggers thermal runaway: a chain reaction where one hot cell heats its neighbor until the whole pack ignites.
To prevent that, every commercial lithium iron phosphate storage system on American grids runs active cooling equipment, including pumps, fans, refrigerants and control electronics, continuously cycling to keep cells in a safe temperature band. That machinery is expensive to buy, expensive to power, and statistically likely to fail before the batteries themselves do. By removing active cooling components, the Wisconsin system’s designer says it eliminates more than 85 percent of the root causes behind historical battery storage failures.
The key is cell formulation. The system relies on sodium ion phosphate pyrophosphate cells, a chemistry that removes cooling systems, eliminates costly routine maintenance and reduces the excess storage required to account for capacity degradation. Sodium ions are larger and less reactive under thermal stress than lithium ions, and that atomic difference is what makes the passive design possible.
The container in eastern Wisconsin and how it got there
The deployment, energized in March 2026, marks the first time sodium ion batteries have backed up the MISO grid, which serves 15 central US states and the Canadian province of Manitoba. The site is a testing facility operated by the energy company that partnered on the pilot, located outside Milwaukee, and the system arrived as a single containerized unit ready to connect.
The developer described it as the largest sodium ion phosphate pyrophosphate battery system in the world and the first grid scale sodium ion storage solution deployed to the US grid. A prior installation at a solar testing facility in Watkins, Colorado stores 3.5 MWh and has been running since 2025, but Watkins was a controlled test environment. Wisconsin is the MISO interconnection, a live market where the battery must respond to real dispatch signals and deliver on contract, not just on a data sheet.
What the performance numbers actually show
Performance testing indicated the passive design could reduce auxiliary power use by up to 90 percent, saving approximately $1 million annually per gigawatt hour installed compared to lithium iron phosphate systems. That figure comes from the developer’s own testing, not an independent audit, and it should be read as an upper bound rather than a guaranteed outcome in every environment.
Even so, a system that draws no power to cool itself will always run at a lower parasitic load than one that does. The developer describes its maintenance schedule as essentially nonexistent, calling the system simple with no moving parts, no planned maintenance and negligible auxiliary loads. As first reported at the time of the pilot agreement, a separate analysis concluded that installing 10 GWh of battery storage across MISO over the next decade could reduce total system costs by as much as $27 billion, and a grid facing surging demand from data centers and electrification is exactly the market where that cost wedge matters most.
Where the technology hits its limits
Sodium ion cells carry less energy per pound than the best lithium ion alternatives, and that gap matters when storage space is tight. The cell at the core of the latest designs achieves around 160 Wh per kilogram, with 97 percent energy conversion efficiency and more than 15,000 cycles at 80 percent capacity retention. Those are competitive figures for stationary storage, but a lithium ion pack targeting maximum density can still beat them on energy per unit of footprint.
Then there is the supply chain. The cells inside the Wisconsin container were manufactured in China to the developer’s specification and assembled in the US. That arrangement is commercially pragmatic today, but it leaves the cost structure exposed to the same trade disruptions that have already rattled lithium supply chains. The answer the developer is building toward is domestic production: a $71 million factory in Sacramento described as the first US facility dedicated to grid scale sodium ion storage, with shipments expected in early 2027, backed by more than 6 GWh of customer commitments.
What the Wisconsin test means for grids built on lithium assumptions
Every major grid operator in the United States has sized its storage procurement pipelines around lithium iron phosphate. The contracts, the interconnection queue, the fire suppression standards and the utility training programs all assume active cooling as baseline. A passively cooled system does not just swap one chemistry for another; it removes an entire mechanical layer the industry spent a decade learning to manage.
The company has secured more than 6 GWh of customer commitments, including a deal with a leading grid storage developer for up to 4.75 GWh through 2030, valued at up to $500 million. Announced customer interest is approaching $1 billion across data centers, critical infrastructure and utility markets. Advances in unconventional grid storage, from a ceramic squeezed to extreme pressure to the sodium ion box in Wisconsin, share the same ambition: get the grid off its dependence on a handful of expensive, fragile minerals. Innovations built on radical physical simplicity, such as pencil-tube generation from raindrops, point toward the same horizon.
The container does not know any of that. It keeps charging and discharging, cooler free, in a state that sells more cheese than it does batteries.
Hugo is an engineer with strong technical expertise. Multilingual from an early age, his writing combines technical clarity with a strong interest in science and energy.
