While Hydrostor Waits for Washington's Check, the Chinese Academy of Sciences Has Already Doubled Compressor Capacity

Last month, Hydrostor CEO Curtis VanWalleghem was asked about the progress of the Willow Rock project at an industry conference. This 500 MW / 4,000 MWh project in Kern County, California, is the largest long-duration energy storage project under development in North America, and had received a $1.76 billion conditional loan commitment from the U.S. Department of Energy's Loan Programs Office (LPO) [1]. VanWalleghem's answer was cautious, saying the new administration was reviewing multiple clean energy financing deals and that everything was in "active communication." In plain English: the money hasn't arrived, and nobody knows when it will. $1.76 billion is not a small number. If that money comes through, Willow Rock will be the first commercial advanced compressed air energy storage plant on U.S. soil.

Around the same time, two things happened on the other side of the Pacific. On January 9, the 300 MW-class compressed air energy storage plant "Nengchu No. 1" in Yingcheng, Hubei Province achieved full-capacity grid connection [2], becoming the world's largest operating plant of its kind. On February 5, the Institute of Engineering Thermophysics at the Chinese Academy of Sciences announced that its jointly developed compressor with China Energy Storage Group had broken the 100 MW single-unit threshold [3], reaching 101 MW with 88.1% efficiency, entirely based on domestically owned intellectual property.

700,000 Cubic Meters Underground in Yingcheng

Yingcheng, Hubei is a salt-producing region with a long mining history, and the ground beneath it is riddled with mined-out salt caverns. The "Nengchu No. 1" project uses these caverns as air storage reservoirs. The principle is straightforward: when electricity is abundant, compressors pump air into underground salt caverns for storage; when power is tight, the high-pressure air is released to drive expansion turbines and generate electricity. Compressed air heats up, and that thermal energy is captured and stored separately, then used to preheat the air during discharge. The working medium is just air. No lithium. No cobalt. The salt cavern walls are dense and pressure-resistant. Thirty years of operation won't produce the kind of capacity degradation you see in lithium batteries. Calling the principle straightforward is accurate at the concept level. Actually converting hundreds of thousands of cubic meters of underground salt caverns into high-pressure vessels capable of repeated charge-discharge cycles is an engineering headache at every turn: cavity stability, sealing integrity requiring ground-up problem-solving.

The underground storage reservoir for "Nengchu No. 1" has a volume of 700,000 cubic meters, with single pressurized spherical tanks 19 meters in diameter, reportedly the world's largest according to China News Service [2]. Total storage capacity is 1,500 MWh, with a daily cycle of 8 hours charging and 5 hours discharging, producing roughly 500 GWh of electricity per year. All key equipment is 100% domestically manufactured. The reason this matters is that there were simply no off-the-shelf suppliers for core compressed air energy storage equipment anywhere in the world. Huntorf and McIntosh used modified conventional gas turbines that were never designed for adiabatic storage. On the Chinese side, everything from compressors to expansion turbines to thermal storage systems had to be developed from scratch. Shaanxi Blower (SBW) makes compressors, Shenyang Blower Works handles expansion turbines, Dongfang Electric does system integration. There are no foreign reference products to buy and reverse-engineer.

Hydrostor's Willow Rock in California has a larger design capacity (500 MW / 4,000 MWh) but is still stuck in the financing stage. They also have a Silver City project in Australia, 200 MW / 1,600 MWh, which their website shows hasn't broken ground [4]. Hydrostor's A-CAES technology has been validated, with a small-scale demonstration running in Ontario, Canada. Goldman Sachs invested $250 million, and a Canadian pension fund also came in. Their investor roster is formidable. The problem is that Willow Rock spent years going from contract signing to land acquisition to environmental review to permitting, and just when they finally secured the LPO loan commitment, the government changed hands in early 2025. The new administration froze review of numerous clean energy financing deals. This isn't just Hydrostor's problem. The entire long-duration energy storage sector in the U.S. is tripped up by the policy cycle. Several companies holding conditional DOE commitments are all standing in the same line, waiting. A UK company called Highview Power is pursuing liquid air energy storage, a different approach that liquefies air rather than compressing it into high-pressure gas, and has built a small demonstration project in Spain. A 324 MW APEX CAES project in Texas has been in planning for years. Every single 300 MW-plus project outside of China remains on paper.

Germany built Huntorf in 1978 and the U.S. built McIntosh in 1991. Both require natural gas to assist expansion, and neither exceeds 50% efficiency. After those two, nobody cracked this path for decades. The core bottleneck: when air is compressed, temperatures rise to several hundred degrees. If that heat is simply dissipated, the air isn't hot enough during expansion to generate electricity efficiently, so you have to burn natural gas to reheat it, which puts you right back on the fossil fuel track. The Institute of Engineering Thermophysics at the Chinese Academy of Sciences started working on this in 2004 [5] and proposed an advanced adiabatic approach in 2009, storing all the heat generated during compression in a thermal storage system and releasing it during expansion to reheat the air, without burning a single cubic meter of natural gas. This approach sounds simple. In practice, it requires solving high-temperature thermal storage material selection, heat exchanger durability under repeated high-temperature high-pressure cycling, and optimization of the coupling between multi-stage compression and multi-stage expansion, each of which is an independent R&D challenge. System efficiency jumped from Huntorf's roughly 42% to above 60%, and was later pushed to around 70%. Then came the 1.5 MW unit in Langfang, the 10 MW units in Bijie and Feicheng, the 100 MW unit in Zhangbei, climbing step by step. China Energy Storage Group is the commercialization company incubated by the research institute, China Energy Engineering Corporation handles EPC, and SBW, Shenyang Blower, and Dongfang Electric make the equipment. Every position in the chain has a corresponding player running the relay. No need for each link to go back to the capital markets to raise another round and find a new company to do the work. As of last year, there were 448 enterprises and research institutions engaged in compressed air energy storage in China. Eighteen salt cavern compressed air energy storage plants were under construction or in planning nationwide, with a cumulative capacity approaching 2,000 MW. China Energy Engineering Corporation released a 660 MW-class system solution [6], with site selection and contracts signed for over 30 plants. The 1,050 MW project in Ulanqab, Inner Mongolia has broken ground, with planned commissioning by the end of next year, which will make it the world's largest. Qianjiang in Hubei is planning five 350 MW-class stations. Heze in Shandong is laying out a 3,060 MW storage base.

Inside the 101 MW Compressor

The 300 MW system at "Nengchu No. 1" uses multiple smaller compressors in parallel. The exact number and individual unit ratings have not been publicly disclosed. Previously, the world's largest single compressed air energy storage compressor was in the roughly 50 MW range. Exact figures vary depending on the source; some literature says over 40 MW, others say just over 50, but none exceeded 60. Based on that range, a 300 MW system would need at least five or six units in parallel. The complexity of piping, controls, and maintenance all scales up accordingly.

Each unit you don't install saves more than just the equipment procurement cost. It saves the associated piping, valves, foundation civil works, control cabinets, and cabling. And every additional unit adds another set of inspection and maintenance schedules downstream. Running multiple units in parallel also creates a problem that non-specialists rarely notice: load distribution and coordinated control between units. When one unit faults or enters maintenance, redistributing load across the remaining units, rebalancing pipeline pressure, all of this makes the control system's complexity grow not linearly but exponentially with the number of units. These costs are not trivial in large-scale energy projects, especially for valves and piping. A single high-quality high-pressure valve can cost hundreds of thousands of dollars. Fewer units means proportionally fewer supporting components, and the savings add up.

88.1% refers to the mechanical efficiency of the compression stage itself: electrical power in, air goes from low pressure to high pressure, how much energy is effectively converted. This number is different from the round-trip electrical efficiency of the entire storage system, which accounts for losses across the full chain of underground storage, heat exchange, and expansion generation. The best publicly reported figures for round-trip efficiency are around 70%. Put in 100 units of electricity, get back 70.

Another parameter, the variable operating range of 38.7% to 118.4%, may deserve more attention than the power doubling itself. The compressor can run continuously from below 40% of rated power all the way to nearly 120% of rated power without shutting down, without frequent start-stop cycling. The primary generation sources paired with compressed air energy storage are wind and solar. A single wind farm can go from near full output to just 10-20% of capacity within a day, then surge back up hours later. If a compressor can only operate within a narrow band of plus or minus 10% around rated power, the moment wind drops, it either shuts down and waits or idles and wastes electricity. Large centrifugal and axial compressors fear frequent start-stop cycling most. Each thermal cycle stresses bearings, seals, and blades with fatigue damage. After a few years, maintenance costs become staggering. A wide variable operating range gives the compressor enough flexibility to follow the temperamental output curves of wind and solar. This parameter is far more useful in real-world operations than nameplate rated power, because the percentage of time a compressor can operate stably under actual conditions directly determines how much electricity a storage plant can charge in a year and how much revenue it generates. If the compressor keeps shutting down every time the wind weakens, utilization hours stay low and the project's financial model doesn't work.

When Salt Caverns Run Out

China's salt mineral resources are concentrated in a handful of provinces: Hubei, Shandong, Jiangsu, and Henan. As it happens, the regions with the highest wind and solar installations and the greatest curtailment pressure, Inner Mongolia, Gansu, and Xinjiang in the northwest, have virtually no usable underground salt caverns. A 300 MW project under construction in Jiuquan, Gansu is experimenting with engineered rock caverns as an alternative, excavating large chambers directly in granite or other hard rock for gas storage. Excavation costs are considerably higher than salt cavern conversion. Industry insiders say privately that costs at least double, and construction timelines are longer. The places that need energy storage most are the very places without natural storage containers. This is the most direct geographical constraint on compressed air energy storage expansion in China.

The Math That Can't Be Avoided

China's infrastructure project completion rate is not as high as the numbers on paper suggest. Of those 448 companies, how many have built something that actually runs and how many registered just to chase local subsidies? Nobody has ever tallied that up. Cases where local governments rush to claim storage projects, cutting corners on geological surveys and economic feasibility studies, number more than one or two. Some salt caverns have actual cavity conditions that differ significantly from design expectations: irregular shapes, interlayers with insoluble material, uneven roof loading. These problems are hard to fully predict on paper. The U.S. side isn't idle either. The DOE's "Long Duration Energy Storage Shot" initiative launched in 2023 [7] targets a 90% reduction in long-duration energy storage costs by 2030. Combined financing from the DOE and private capital for Hydrostor, Form Energy, and others is not insignificant. It's just that every step from financing to permits to supply chain buildout is constrained by political cycles and capital markets' short-term preferences. Industrial giants like Siemens Energy and Ingersoll Rand are also positioning themselves, though their level of commitment looks more like "staking a claim and watching the market."

A 70% round-trip efficiency compared to lithium's 90%+ is a 20-percentage-point gap. Putting those two numbers side by side isn't entirely fair. Lithium batteries are efficient and fast-responding for two-to-four-hour short-duration storage. After a few thousand cycles, capacity starts dropping. You're looking at replacement after ten or so years. There's fire risk. Nobody has come up with a fire protection design for large containerized battery storage plants that's truly reassuring. Last year alone, several storage plant fires in China made the news. And then there's the supply chain pressure on lithium, cobalt, and nickel. These issues are tolerable in a two-to-three-hour peaking scenario. Compressed air energy storage is positioned for eight-plus-hour long-duration storage, with single-plant capacities at the gigawatt-hour scale and thirty-year lifespans. The 70% round-trip efficiency is improving, but the ceiling is there. The second law of thermodynamics is non-negotiable. Irreversible losses in gas compression-expansion cycles can only be squeezed so much further. The 101 MW compressor reduces system costs by cutting the number of units needed. The 660 MW-class standardized solution reduces engineering costs by eliminating redundant design work. These are all cost-reduction levers. The entire industry still runs heavily on policy. Local government storage quotas, central subsidy levels, ancillary service pricing mechanisms in electricity markets: these policy variables determine whether a project gets built. The true market-driven inflection point hasn't arrived. Many projects getting built today owe their existence to local governments mandating that new renewable energy projects include a certain percentage of co-located storage. Developers must build storage plants to secure grid connection permits for their wind and solar projects. Whether the storage plant itself turns a profit is sometimes not the first consideration. Sodium-ion batteries, flow batteries, and other long-duration storage technologies are also gaining ground.

Insurers didn't know how to price the risk. Banks couldn't assess their exposure. Those who had already invested felt uneasy. Those thinking about investing stayed on the sidelines. Under the current U.S. policy environment, convincing investors that an infrastructure project with a ten-year payback period will definitely be profitable a decade from now is a tall order. Under the Chinese policy environment, the money shows up and the projects get pushed through. But when the standards aren't in place, who underwrites thirty years of operational lifespan? That question has no answer either.

Based on publicly available information. For industry reference only. Does not constitute investment advice.