Solid-state transformers could ease the grid transformer shortage

Transformers are a vital but little-noticed part of the grid. They step voltage up for long-distance transmission and down for homes, and their basic design has barely changed in about a century. They are also in short supply. As demand rises from EVs, data centers and more, researchers and companies are trying to modernize them, and MIT Technology Review looks at the solid-state version.

A conventional transformer uses a core of grain-oriented electrical steel with coils of copper wire at either end. The ratio of loops in the two coils sets the voltage change, and magnetic induction carries the energy across. Transformers also often provide galvanic isolation, meaning there is no direct electrical path through them, which helps keep disturbances from spreading through the grid. Srdjan Lukic, a professor of electrical and computer engineering at North Carolina State University, says they form the backbone of transmission and distribution, but today's units are big, bulky, and difficult and expensive to replace.

A solid-state transformer adds power electronics, meaning semiconductors that convert and control electricity, to that basic blueprint. Where a conventional unit runs at the frequency of the grid it is attached to, a solid-state unit controls the frequency itself. It typically converts incoming AC to DC, then to higher-frequency AC using switches made of a semiconductor such as silicon carbide. A magnetic component then changes the voltage, and because the electronics give so much control over the current, that core can be much smaller. Lukic says some come in at roughly one-tenth the size at the same capacity. More power electronics then convert the output back to DC and finally to lower-frequency AC. Because these devices absorb the jobs of rectifiers and inverters, they can shrink the total footprint of the equipment and potentially cut power lost in the system.

In August, the New York Power Authority and the Electric Power Research Institute (EPRI) tested a one-megawatt solid-state transformer under realistic utility conditions at EPRI's test site. The device was built in Lukic's lab, and he says it was the first such test for a transformer of that capacity. The prototype was roughly 98% efficient, losing only 2% of the energy fed in. Lukic says an equivalent conventional transformer and rectifier would be about 96% efficient together, so the solid-state unit could cut losses from conventional equipment in half. A main goal was to give utilities more confidence in the new components. The prototype is still a relatively low-voltage system, at about 15,000 volts. Industrial facilities such as larger data centers often need higher-voltage connections, and transmission lines typically run at hundreds of thousands of volts, so adapting the design will take further engineering. Lukic calls the test a successful demonstration that solid-state transformers can work under real conditions.

For now, companies in the field are largely focused on data centers, Lukic says, because money is flowing in and the voltage requirements are lower. In the near term, solid-state transformers could be used inside data centers to provide electricity for server racks.

The article profiles one company, Heron Power. In February it raised $140 million to commercialize its solid-state transformer, which it calls the Heron Link. Founder and CEO Drew Baglino says the technology could serve data centers, solar arrays and grid batteries. Heron is starting to test at larger scales: ten megawatt-hours of batteries at its Scotts Valley, California, headquarters will soon be tested with Heron transformers, and it is planning a pilot with a nearby solar farm. It is also building a $100 million factory in California, which should begin production in 2027 and reach a full rate of 40 gigawatts of transformers per year in 2029.

One known weakness is that the power electronics have more failure modes than a conventional transformer, where a failure typically takes the whole device offline and interrupts power. Baglino says Heron uses a modular design with many small devices working together, so if one fails the rest keeps running until the affected part is replaced, without shutting everything down.

The shortage is the backdrop. Benjamin Boucher, a supply chain analyst at Wood Mackenzie, says grid equipment makers have recently put $1.7 billion into the US supply chain for power transformers, which are used in high-voltage transmission, with another $1.3 billion for distribution transformers, the smaller units serving homes and businesses. The article says even this is not keeping pace with the grid's growth, and wait times for large conventional transformers are stretching as long as five years. Boucher says transformers have become one of the biggest bottlenecks for utilities, generation developers and even data centers, and that new manufacturing is hard to start because the process is technical and needs a lot of specialized labor. Facilities announced years ago are still years from opening.

Solid-state transformers could help ease some of that pressure, the article says. But the high-voltage devices needed for power plant connections and transmission lines are the holy grail, Lukic says. Existing solid-state technology cannot be used for those applications yet, and the industry needs new materials and designs to handle high voltages safely in a compact device.

Key facts

In August, the New York Power Authority and EPRI tested a one-megawatt solid-state transformer built in Srdjan Lukic's lab; Lukic says it was the first such test at that capacity.
Lukic says the prototype was roughly 98% efficient, against about 96% for an equivalent conventional transformer and rectifier together, which he says could cut losses in half.
The prototype runs at about 15,000 volts; high-voltage versions for power plants and transmission lines still need new materials and designs.
Heron Power raised $140 million in February for its Heron Link and is building a $100 million California factory, due to start production in 2027 and reach 40 gigawatts a year in 2029.
Wait times for large conventional transformers are stretching as long as five years, even after $1.7 billion and $1.3 billion of recent US supply chain investment, per Wood Mackenzie's Benjamin Boucher.

Why it matters

Transformers are the backbone of electricity transmission and distribution, and their basic design is about a century old. The article says they are now a bottleneck: wait times for large conventional units stretch as long as five years, and Wood Mackenzie's Benjamin Boucher says transformers have become one of the biggest bottlenecks for utilities, generation developers and even data centers. Recent investment of $1.7 billion for power transformers and $1.3 billion for distribution transformers is not keeping pace with the grid's growth. Solid-state designs use semiconductors and a much smaller magnetic core, and they take over the work of rectifiers and inverters, so they offer a different route to the same job. The article's claim is modest: they could help ease some of that pressure.

Who it affects

Utilities, generation developers and data center operators are the groups Boucher names as squeezed by the shortage. Data centers are also where the technology is headed first: Lukic says developers are largely focused on them because money is flowing in and voltage requirements are lower. Heron's Baglino also points to solar arrays and grid batteries. Longer term, anyone who relies on high-voltage transmission would be affected if the technology matures, but the article says it cannot serve those uses yet.

How to use it

There is nothing to buy or deploy off the shelf in this article. In the near term, the article says solid-state transformers could be used within data centers to provide electricity for server racks. Heron Power is moving toward products: ten megawatt-hours of batteries at its Scotts Valley headquarters will soon be tested with its transformers, a pilot with a nearby solar farm is planned, and its factory should begin production in 2027. The NYPA and EPRI test was aimed at giving utilities more confidence in these components before any grid use.

How solid is it

The article is an explainer from MIT Technology Review that draws on named sources: Lukic of North Carolina State University, Heron's founder and CEO Drew Baglino, and Boucher of Wood Mackenzie. The headline efficiency figures (roughly 98% against about 96%) come from Lukic, whose lab built the device; no separate figures from EPRI or NYPA are quoted. The one-megawatt test is described as successful and as the first of its kind at that capacity, again per Lukic. Heron's production dates are stated as what the facility should do, not as achieved results.

Risks and caveats

The technology is still in development and cannot replace all conventional transformers today, especially the largest ones that work at the high voltages needed for long-distance transmission. The prototype operates at about 15,000 volts, and adapting it to higher voltages will need more engineering. The article reports no price or cost comparison with conventional units. Solid-state units also have more failure modes because of their power electronics; Heron's answer is a modular design in which the rest of the system keeps working if one part fails. Lukic's claim that losses could be cut in half rests on one prototype and one comparison. Boucher also notes that new transformer manufacturing is hard to get running, so even announced facilities are years from opening.

“Without the transformers, you don’t really have an electrical grid.”

— Benjamin Boucher, supply chain analyst at Wood Mackenzie