Scaling GETs may be harder than DOE has painted it

In April, the Department of Energy’s Loan Programs Office released the tenth installment of a series of reports aimed at jump-starting investments in certain critical technologies. The 20-odd solutions known collectively as “grid enhancing technologies” or GETs have the potential, DOE said, to support up to 100 gigawatts of incremental peak demand.

The perhaps-ambitious deployment timeline? New grid tech could increase capacity by up to 100 gigawatts if installed “overnight," and scale up over the next three to five years.

Sila Kiliccote, an investor at Breakthrough Energy Ventures, told Latitude Media that while investors will do “everything we can,” she’s a little skeptical that such a condensed timeline is realistic.

There’s plenty of interest and momentum in GETs from an innovation and investment perspective, Kiliccote said. But in her experience of taking tech from lab to market, it’s likely to be difficult to push adoption that quickly if the market traction isn’t already in place.

“There has to be markets and politics to support it; there has to be pull from utilities; there has to be policy support to get these technologies off into the world,” she said. ”And unless they come together in a very synced-in way, three years is a very ambitious goal.”

That said, the industry behind the energy scramble that’s creating a more urgent need for GETs could also compel their proliferation, Kiliccote said.

“Data centers open up a huge customer opportunity [for GETs],” she said. “They can be the first customers for some of these technologies; they can be the only customers for some.”

Starting with data centers

One of the primary challenges facing GETs startups is that a lot of investors will have concerns about a go-to-market strategy that relies solely on utilities, Kiliccote explained.

Selling into that sector takes a very long time, is extremely competitive, and usually involves a request for proposal (RFP) as opposed to a direct purchase.

Because of that, many startups look to other sectors where they can establish “meaningful revenue” before delving into the utility sector, even if that’s where they eventually hope to make the biggest impact.

“Some of these technologies could really be smaller or shorter versions of the ones that are going to be adopted for the electricity grid,” she added.

For some technologies, advanced conductors for example, the data center sector is an attractive proving ground. They can be deployed to bring power from renewables to data centers themselves, but also provide lower voltage power distribution within a data center, Kiliccote said.

A company’s first addressable market is a key element of how Kiliccote and Breakthrough assess potential investments. 

“When we invest in companies, we really look at what the opportunities are in their first markets, and what it would look like in terms of impact if they stayed there, and never made it to the utility market,” she explained. “If they stay there, could it still be a big impactful company?”

The data center market may be a relatively solid bet on those terms, she said, because it’s growing so quickly.

But ultimately, success really depends on the tech itself: how easily it can be scaled up or down, how it could perform behind the meter, and at various voltage systems. Power electronics startups, for example, probably really want to go straight into the utility market, given the small size of other applications.

Getting comfortable with GETs

One way to accelerate the deployment of GETs, Kiliccote said, could be to develop testing standards.

For instance, conductors don’t currently have those standards, which means that a company can’t send its tech out to a lab to get it certified. And that leaves individual utilities doing their own testing. That customer-by-customer approach to verification makes it really difficult to build and sell any product, she added.

That’s a gap Kiliccote and her fellow investors have seen in conversations with GETs companies more broadly, she said: “They’re trying to figure out what each utility or customer wants, they are developing their own metrics, and they’re trying to develop their own testing.”

It’s a bit of a “chicken and egg problem,” she added. “Who's going to test it? Who's going to certify it? Who's going to verify that it works the way it should work? Having some comfort around these is very important.”

Where data centers serve as first markets for GETs, there needs to be an effort to translate reliability, safety, and security lessons into the utility world, Kiliccote said: “If data centers use these technologies and don’t communicate those benefits in a way that’s transferable to the power sector, they may only be used in the data center world.”

That said, those lines of communication are, in many cases, already open, Kiliccote added. And ultimately, she is “still optimistic” on GETs deployment at scale on a relatively rapid timeline.

“With the attention to AI, the attention and the needs of data centers, I feel like the inflection point may come in the next three to five years,” she said. “When it takes off, it’s going to take off really fast. And those who are in the field with the product are going to be surprised by how fast they’re going to need to answer this demand.”

Editor's note: This story was updated with a correction on June 26. The DOE report did not say scaling up GETs is possible in one to three years, but rather that deployment is immediately possible and scale is achievable in three to five years.