The Michigan-based battery energy storage technology startup Volt Harbor has raised $2 million in seed money. The company makes a modular, software-defined energy storage platform for data centers and the grid. Volt Harbor’s technology works with EV batteries, which are in abundant supply and are growing in quantity. When electric vehicle battery packs can no longer be used for operating electric vehicles, they still have quite a bit of energy capacity remaining to be utilized for second-life stationary energy storage purposes. As far back as ten years ago, some used Nissan LEAF batteries were re-purposed for use in a French data center. Since then, there has been an explosion in EV manufacturing and sales. Consequently, the number of used EV batteries available has grown tremendously. Dr. Al-Thaddeus Avestruz, President, CEO, and co-founder of Volt Harbor, answered some questions for CleanTechnica. How does your technology work more efficiently with converters? Conventional battery energy storage systems process 100% of the energy that comes out of the batteries. That works, but converters are expensive, and they waste energy every time power flows through them. So the larger or the more converters you have, the more cost and inefficiency build up across the system. We mathematically modeled the variation observed across used and mixed batteries. Some packs are weaker, some are stronger, some have different chemistries. We found that you don’t need to process all of the power from the battery. You need the right converters in the right places, along with the right software architecture — our patented Media Access Control (MAC) approach — to let battery modules from different manufacturers share with small power converters coordinating power flow through the system in real time. We use fewer converters often arranged in two coordinated layers: a small number of higher-power converters handle the average mismatches across groups of batteries, and a larger number of smaller, lower-cost converters handle the residual individual variations. The exemplary result, validated in peer-reviewed research, is 94% energy utilization compared to about 78% for conventional partial-power processing and 23% for full-power (100%) processing: meaning more of the battery’s actual energy reaches the application, with significantly lower hardware costs. In addition, our architecture creates AC directly from the batteries. This eliminates inefficiency and cost of a DC-AC PCS (inverter Power Conversion System) conversion step. Overall, our system has 10-20% the cost of power conversion compared to conventional systems. What capacity is your pilot energy storage system with DTE Energy? Our initial pilot system with DTE is a 100 kW / 300 kWh deployment designed to buffer high-power EV charging events and demonstrate the platform’s multi-use capabilities, including demand response, peak shaving, and backup support. While the demonstration size reflects the goals of the Emerging Technology Fund partnership, the MAC-BESS™ architecture is engineered to scale modularly, from compact commercial deployments to multi-megawatt utility-scale systems. What are the advantages your energy storage system has? A few that matter most for grid and commercial operators: Lower total system cost. Because the architecture uses fewer power converters, capital costs are materially lower than those of conventional BESS approaches, particularly important for cost-sensitive utility-scale and C&I deployments. For second-life applications specifically, we deliver storage at roughly one-half to one-third the cost of new battery systems with equivalent capability. Battery-source flexibility. Our software-defined energy systems tolerate heterogeneity at the power-electronics layer, enabling us to integrate battery modules from multiple OEMs and chemistries — lithium-ion, LFP, NMC, sodium-ion, etc. — into the same system. That eliminates the need to sort and homogenize incoming battery supply, which is typically the most expensive part of running a second-life operation. Extended battery life. Our software-defined battery management approach can extend usable battery life by up to 30% compared to conventional approaches, further improving project economics. Sub-100-microsecond response time. That’s fast enough to ride through grid disturbances that would otherwise trigger backup generation: important for data center and critical-load applications. Reliability comparable to aerospace and commercial aviation. Single parts-per-million failure rates and no single point of failure, because the architecture is distributed across modules rather than dependent on a single controller, converter, or battery. Each field-replaceable unit is hot-swappable with no interruption of operation. Modular scaling. The same building blocks scale from commercial-sized deployments to multi-megawatt utility-scale and data center systems. What EV battery packs can it work with? The platform is designed to work with battery modules from a wide range of EV manufacturers and chemistries. This includes lithium-ion (NMC, NCA) and lithium iron phosphate (LFP), with the architecture also extensible to emerging chemistries like sodium-ion as those enter the market. A key design choice is that we work at the module level rather than requiring intact OEM packs. This means we can optimize our operation for a wide range of voltage requirements. That means batteries from different vehicles, different model years, and different manufacturers can be integrated into the same system at a high yield without needing to be sorted, homogenized, or kept matched by chemistry. The power-electronics architecture handles heterogeneity. As the supply of retired EV batteries grows and diversifies over the coming decade, this flexibility becomes increasingly important. It means second-life supply chains don’t need to rely on exclusive OEM deals or large, presorted inventories to operate economically. Can your system work with data centers? Yes. The data center market is a primary near-term focus alongside utility and C&I deployments. The same MAC-BESS™ architecture that handles heterogeneous second-life batteries on the cost-sensitive side is also engineered to handle the reliability and performance demands of data center power infrastructure on the high-end side using new batteries. A few things make MAC-BESS™ a particularly strong fit for data centers: The sub-100-microsecond energy delivery is fast enough to ride through load and grid fluctuations that would otherwise cause failures or trigger backup generation. With AI workloads driving rapid power swings, sometimes from 30% to 100% of rated capacity within seconds, traditional data center power systems weren’t designed for this kind of volatility. The single parts-per-million failure rates and the absence of any single point of failure meet the reliability standards that data center operators expect for critical loads. And the integrated, software-defined architecture is designed to reduce the number of discrete systems data center operators need to specify, integrate, and maintain, meaningful in an environment where grid connection wait times in major data center markets now exceed four years, and operators are looking for faster, more configurable paths to bringing capacity online. Some people think the cost of batteries will continue to decline… do you foresee used EV batteries will still be viable for repurposing into stationary energy storage for some time? This is a question we think carefully about, and our view is that second life will remain economically viable for the foreseeable future, but the durability of our business to work with both new and second-life batteries doesn’t depend on it. On the second-life question specifically: even with new battery prices declining, the cost gap between new and second-life batteries remains substantial. Our platform delivers second-life storage at one-half to one-third the cost of new equivalents, and that economic delta is durable because second-life batteries also avoid the embedded carbon, mineral extraction, and manufacturing costs that new cells carry. As ESG and lifecycle-emissions requirements continue to tighten for utilities, data center operators, and commercial buyers, those non-price factors increasingly matter to procurement decisions. There’s also a supply consideration that often gets overlooked. Industry analysts project that the supply of retired EV batteries will grow more than 10x by 2030. Even if new battery prices keep falling, that’s an enormous volume of usable capacity that will exist regardless. Shredding all of it for materials recovery means burning the plastics, insulation, and other organics, plus losing 70-80% of the remaining energy capacity that’s already been manufactured. Putting those batteries back to work in stationary applications is a better environmental and economic outcome. But the more important point: Volt Harbor’s platform is designed to work with both new and second-life batteries. Our value proposition is the architecture itself. Its software-defined, and modular, capable of handling heterogeneous battery sources, with the lowest cost of power electronics. It’s not a bet on which battery supply ends up cheapest. In fact, as the cost of batteries decreases, power electronics become a larger fraction of the system cost. As the cost dynamics evolve, our customers can use whichever supply makes sense for their application, and our system works either way. That flexibility is part of why the architecture is durable across whatever market shifts come next.
