The grid is being reshaped not by a single top-down plan but by millions of individual decisions. Customers at every scale — a restaurant in New York City plugging in a battery, a commercial building owner running the numbers on payback, a multi-gigawatt AI data center racing to get online — are investing in on-site, intelligent, interactive energy resources. Individually they’re chasing resilience, savings, and speed to power. Collectively, they’re expanding the grid into territory it never reached and forcing a long-overdue conversation about how the system is governed.
This live panel, recorded July 8, 2026, to open The Great Transformation conference, brings together a policy advocate, a global equipment and controls provider, and a building-energy software founder to map that shift. The through-line: this transformation is already happening and can’t be stopped, so the real question is how we accelerate it, keep it from fragmenting the grid, and make sure the benefits reach the communities where these resources are built.
I interview three people who aren’t just watching this shift — they’re shaping it: Cameron Brooks of Think Microgrid, Greg Castle of Schneider Electric, and Mike Grenier of Elexity.
The panel digs into the data center power boom (a reported 114 GW of behind-the-meter generation in the pipeline — roughly 20% of U.S. peak load), why falling battery prices are flipping “two yeses out of ten” into “eight,” the “Frankenstein microgrid” phenomenon, and how flexible interconnection deals in New Jersey and the Google–Voltus capacity agreement point toward a grid that’s built from the outside in. It’s a candid, optimistic conversation about ownership, resilience, and scaling what already works.
In this episode you’ll hear:
• Why 114 GW of behind-the-meter data center power — France and the UK combined — is being planned in the next five years
• How a payback period dropping from seven years to three flips customer decisions from “no” to “yes”
• What a “Frankenstein microgrid” is, and why the software and controls layer is now the hard part
• How New Jersey and the Google–Voltus deal use distributed batteries to unlock grid capacity faster
• Why the panel frames utility economics — not technology — as the airway that has to be opened first
Guests
Cameron Brooks — Executive Director, Think Microgrid
Cameron Brooks leads Think Microgrid, a coalition that serves as a unified policy and regulatory voice for the U.S. microgrid industry. Founded in 2021 in partnership with Microgrid Knowledge, the organization advocates for microgrids in legislative and regulatory proceedings at the state and federal level. Brooks is also president of E9 Insight, a research firm focused on the U.S. utility industry and its evolving policy landscape, and has held leadership roles across clean-energy companies and non-profits. In this conversation he argues that the shift to on-site resources is less a “transformation” than an “expansion” into places the grid never reached — and that upside-down utility economics are the first thing that has to change.
Greg Castle — Commercial Leader, North American Power & Grid, Schneider Electric
Greg Castle leads the commercial side of Schneider Electric’s North American power and grid segment, working directly on large-scale data center power projects. He brings the on-the-ground view: gigawatt-scale builds that now reach five to ten gigawatts, treated increasingly like privatized utilities and managed with grid-scale ADMS/DERMS control layers. Castle is the source of the episode’s standout term — the “Frankenstein microgrid,” pieced together from whatever generation developers could buy during the supply-chain scramble.
Mike Grenier — President, Elexity
Mike Grenier is president of Elexity (formerly Extensible Energy), a Bend, Oregon–based company whose software optimizes energy assets — storage, solar, HVAC, and EV charging — in commercial buildings. Talking daily with commercial building owners, Grenier offers the customer-economics lens: the predictable “magic point” where a payback period crossing from seven to three years unlocks mass adoption, and why clear price signals from utilities could accelerate everything.
Timestamped Topic Map
00:00
Cold open and series intro — the grid is undergoing massive change, creating opportunity for local energy
00:30
Elisa Wood frames the episode: customers investing in distributed energy are collectively reshaping the system
00:50
Setting: recorded live to open The Great Transformation conference, Bend, Oregon, July 8
01:20
Meet the panel — Cameron Brooks, Greg Castle, Mike Grenier
01:30
Greg: ~114 GW of behind-the-meter data center generation in the pipeline (Bloomberg)
02:00
Scale check — that’s ~20% of U.S. peak load, roughly France and the UK combined
02:40
Cameron introduces Think Microgrid; reframes “transformation” as “expansion” into new territory
03:20
Three drivers of on-site investment: resilience, economics, and speed to power
04:40
Cameron: this is a good thing — build policy for a grid that grows “from the outside in”
06:10
Mike: how small economic changes unlock massive customer investment
06:40
The “magic point” — payback dropping 7 → 5 → 3 years flips 2 yeses into 8
08:00
Greg on the ground: 5–10 GW projects; treating campuses as privatized utilities; ADMS/DERMS
09:00
The “most violent load profile anyone has ever seen” and three-to-five-nines reliability
10:00
Utility side: faster uptake of VPPs, demand response, distributed storage — but a scale problem
11:00
Elisa presses the panel: how do we scale, and what’s in the way?
11:40
Cameron: it’s happening at every scale; NYC restaurants use plug-in storage, no interconnection
13:30
Utility economics are “upside down” — incentivized to invest capital, not cut costs
14:20
The EMT metaphor: fix the “airway” (utility economics) or nothing else matters
15:30
Greg: New Jersey lets data centers interconnect faster by funding residential batteries (a VPP)
16:40
Mike reframes: it’s a snowball — the question is how to accelerate via clear price signals
17:40
Elisa: data centers have a community-opposition problem; energy has the solution
18:30
Newsletter break — the Energy Changemakers weekly briefing
19:30
Audience Q (equity): how do we scale without giving all the money to those already benefiting?
20:20
Cameron: ownership vs. “renting” our energy; keeping benefits local; Oregon microgrid zones
22:40
Audience Q (Jake Edie, RenewComm): the absence of interconnection & siting barriers on-site
23:00
Mike: large generation takes 6–10 years; a solar + battery install took three months
23:30
Greg: interconnection queues push data centers off-grid; most would connect if they could
24:30
Cameron: a fully fragmented grid is the worst outcome — invite interconnection instead
25:30
Cameron’s five-day December outage; resilience becomes non-negotiable; reasons for optimism
27:00
Back to the economics of data centers
27:20
Mike: data centers spotlight the capacity problem; the Google–Voltus “capacity” deal
28:30
Greg: average U.S. grid utilization is ~52%; built for one or two peak days a year
29:20
Cameron: two constraints — capacity and peak load; don’t brand data centers as defectors
30:40
Virginia’s regulators wrestle with how to incentivize flexibility
31:00
Three kinds of interconnection: physical, data/information, and economic
31:40
The “dirty little secret”: data centers themselves aren’t flexible — partnerships bring flexibility
32:40
Award for best new term: Greg Castle’s “Frankenstein microgrid”
33:20
The gold rush for generation; turbines unavailable until 2033; why controls are the hard part
35:30
Audience Q (Kat, Clackamas Community College): advice for students entering energy
36:00
Mike: learn fast, communicate clearly — AI makes those skills more valuable, not less
36:50
Greg: an exciting time; AI is both cause and cure; everything is shifting to DC (800 VDC)
38:20
Cameron’s three pieces of advice: policy matters, follow the physics, learn from telecom
41:00
Elisa thanks the panel and organizers Robert Cross and Skip Newberry; outro
Transcript (AI derived)
ENERGY CHANGEMAKERS PODCAST
Episode 53 — Corrected Transcript
How Customers Are Reshaping the Grid
Host
Elisa Wood
Guests
Cameron Brooks (Think Microgrid), Greg Castle (Schneider Electric), Mike Grenier (Elexity)
Recorded
Live — The Great Transformation, Bend, Oregon, July 8, 2026
Editing
Filler words removed and false starts cleaned for readability; wording preserved. Proper nouns verified and corrected. See corrections log at end.
Transcript
Introduction
Announcer [00:00]
We know the grid is undergoing massive change, which creates massive opportunities for local energy. The Energy Changemakers podcast offers rich conversations with industry leaders to help you and your company capture opportunities to build a decentralized grid. I’m your host, Elisa Wood. Welcome to the Energy Changemakers podcast.
Elisa Wood [00:20]
Hey everyone, Elisa Wood here from Energy Changemakers. These days, we’re seeing more and more customers — from homeowners to businesses to AI data centers — investing in distributed energy. They’re doing so for their own purposes, but collectively, they’re doing something much bigger: reshaping how the electric system itself evolves.
I explored this idea at The Great Transformation conference in Bend, Oregon on July 8th. The three-day event about distributed energy was sponsored by Outfit and the Technology Association of Oregon. We opened the event with this live podcast. I love this conversation because these folks aren’t just following these changes — they’re helping shape them.
You’re going to hear from Cameron Brooks, executive director of Think Microgrid; Greg Castle, North American commercial leader for power and grid at Schneider Electric; and Mike Grenier, president of Elexity. My first question to the panel was: what’s the biggest shift happening in the power sector now? Greg kicks it off with some pretty astonishing data on the growth of behind-the-meter power for data centers. Here we go.
The biggest shift: behind-the-meter power
Greg Castle [01:30]
Rapidly, it has moved to having more of the on-site power generation for data centers. And again, if you haven’t seen the Bloomberg article, they quantified roughly 114 gigawatts worth of behind-the-meter power generation that’s either in development or being planned to be built in the next five years.
Just for context, that’s 20% of the entire peak load of the United States, and it would basically be the equivalent of France and the UK combined — and that is behind-the-meter power generation. It’s unfortunate, and we’ll talk a little more about public sentiment with data centers, but I think we’re only going to see that trajectory grow, which will have some negative ramifications from a community standpoint.
Schneider Electric is pretty heavily involved with the data center world, and it’s just been really astonishing. It was like, overnight, the on-site power generation — bridge power, or even completely off-grid — is completely transforming the energy markets.
Cameron Brooks [02:40]
Great, thank you. I know Elisa already said my name, but I’m Cameron Brooks. I’m the executive director of Think Microgrid. Maybe just a little context: Think Microgrid is an organization that serves as the unified voice of the microgrid industry. We’ve been around since 2021, and we were actually founded in partnership with Microgrid Knowledge — very grateful for the incubation we had there to be able to grow the organization.
To come back to what people are not seeing or not understanding: I think both of you have touched on the economics and the shift to on-site resources. And to go back to the idea of transformation — in a way, I’m not sure it’s a transformation as much as it’s an expansion into areas the grid just hasn’t reached before.
What I’m seeing is that companies, customers, and communities at all scales are investing in resources that are on-site, smart, intelligent, and interactive, for a few different reasons. One is resilience. People are looking for backup power, and suddenly backup power can be way more than backup power that kicks on only when the power goes out — it can be a battery system that’s there all year long, twenty-four hours a day.
They’re looking at economics and seeing that on-site resources and microgrids can be not just a cheaper solution, but a hedge against what is only going to be an increasing cost trajectory for utility-delivered power. And then we hear a lot about speed to power, which is maybe where microgrids started, in terms of just having energy access. It started in remote communities, but data centers are really changing the equation: we want power now, and the grid can’t deliver it, so we’re going to build our own resources.
The thing that I don’t know if people are getting wrong, but that I’d invite people to reframe, is this: this is all a good thing. This should be celebrated. We should be looking at these resources not as something the community doesn’t want. There are a lot of issues around data centers and energy development that do require careful attention — water, land use, all kinds of things. But the idea of having resources on-site that can be interactive, supporting the traditional grid while also providing things the traditional grid cannot provide — that’s worth celebrating.
The traditional grid cannot provide reliable backup power that’s on-site. That has to be on-site. And when we have a grid where a wildfire or a squirrel can take it down, then you want to have something on-site. So again, I’d say this is something we should look at as something we want, and think about how we build a regulatory, policy, and economic framework around the idea that the grid is not just a central system delivering power, but that it’s actually coming from the outside in as well — and that’s something we want to plan for, not something we want to be dragged into kicking and screaming, because there’s a lot of opportunity. So that’s what I’d offer.
Elisa Wood [06:00]
Thank you, Cameron. It’s an amazing time we’re in right now. We’re in a kind of renaissance, and we don’t necessarily even realize it, but we will realize it very soon. I want to turn to Mike now. Mike, we had a conversation recently, and you were describing how relatively small economic changes can suddenly unlock massive customer investment. What does that mean? Explain that to us.
The economics: the payback “magic point”
Mike Grenier [06:20]
What it really tells is a story. We spend every day talking to commercial building owners, because our software manages energy for commercial buildings. When we talk to customers about investing in their energy assets, the first thing they ask is, “What’s in it for me?” And the answer you have to give the CFO is, “This is your payback,” or, “This is your IRR.”
There is a very predictable magic point when the payback goes from seven years to five years to three years, where all of a sudden it goes from two people saying yes out of ten to eight people saying yes out of ten. As those costs come down for these projects, those payback periods are dropping from seven to five to three. And we’re seeing that right now, out in the market. We’re talking to building owners, and just that little change in battery prices in the last two years has switched them from two yeses to eight yeses.
Elisa Wood [07:40]
It’s amazing, because we have this time when it’s becoming much more economically feasible to do on-site energy because of these battery prices. So I want to turn to Greg, because Greg, you are seeing this actually play out — you’re doing actual projects. What does this transformation look like on the ground today when you’re out there talking to data centers and other customers?
Greg Castle [08:00]
Yeah, I’ll double down on the data center power generation side. What is very different is the scale and the pace of some of these projects. It used to be unheard of that you would see a gigawatt project, and now we’re seeing five- and ten-gigawatt projects come across our desk. We’ve really had to reimagine and re-architect how we look at these projects.
Schneider’s approach, at least on the larger ones, is to look at them more as privatized utilities. We happen to have a world-class digital grid portfolio for ADMS and DERMS, both in front of the meter and behind the meter. So we’ve been able to cherry-pick certain aspects of our utility portfolio, along with some other control layers, to help manage the complexity of these projects — because you’re talking multiple gigawatts of very diverse assets: recips, aeroderivatives, sometimes turbines, along with large-scale BESS. And then you’re trying to pair that with the most violent load profile anyone has ever seen, with data centers.
And then, eventually, they want to interconnect into the grid — maybe that’s three, five, or seven years down the road. So the energy management system and the controls for the campus have become very critical, not only to protect the equipment from the data center, but to drive the three-nines, four-nines, five-nines reliability, and also to enable flexibility both to the data center and potentially to grid services down the road.
And it’s not even just one of these projects. There are forty or fifty of these large-scale projects being built, and they have the intention to grid-connect down the road — which we hope they do, because the grid needs it, but they don’t necessarily have to. To flip the equation to the utility side: what we’ve noticed is a more rapid increase in adoption of certain technologies, like VPPs and ramping up demand response, and looking more at distributed storage in communities and neighborhoods at scale. But the challenge is that we’ve been talking about this for ten or fifteen years, and it’s not a technology problem — it’s a scale problem. So the question is, how do we scale faster? This really needs to be done in the next three to five years for us to make a positive impact and innovate on the grid.
How do we scale?
Elisa Wood [11:00]
Okay, we’re going to get into the economics of data centers a little bit, but first I have to jump on your last point and follow up with the three of you, because that is key. You made such a good point — how do we scale? Who wants to take that? What do we need here? What’s getting in our way? What’s your pain point when you’re trying to work with customers — is it lack of education, regulatory issues, inability to get financing? As you said, we’ve been talking about this for ten or fifteen years. Some people in this room have been talking about it longer than that. Now is the time, the need is there, the customer is there, but on-site power is still pretty niche. It’s getting there and changing, but still pretty niche. So what do we do?
Cameron Brooks [11:40]
Well, I’ll take that. I don’t know that I have an exact answer for you, Elisa, but I will say that when we talk about scale, one of the things I think is most interesting is that it’s actually happening at every scale of the electric industry — from devices to appliances to homes to commercial buildings on up to data centers.
Part of what’s driving that change, clearly, is storage and the cost trajectories around having storage available, and having on-site resources — solar, fuel cells, what have you — that can feed into it. So when we talk about scale, it’s happening at every single step of the way. It’s creeping up on us, and I think we’re going to get to that tipping point very quickly. We’re already seeing it. Plug-in storage is available now. Restaurants in New York City are using plug-in storage with no interconnection required and no red tape, to manage the costs they’re facing against their utility.
Here in the Intermountain West, more and more homes — and businesses and communities as well — are looking at opportunities to have storage as a resilience strategy, because so far the utility industry seems to only know how to bury power lines, and that’s not really a complete solution.
And that leads me into part of the answer. One of the challenges we face very clearly is that the economics of the utility industry are completely upside down. Unlike most industries — and this is another thing people outside this room and outside our world don’t understand — they think that, like any other industry, utilities are looking for ways to cut their costs, be more efficient and effective, and bring a product to customers at a lower cost. Well, that’s exactly the opposite of what the utility industry, at least the investor-owned utility industry that has a gravitational pull on the rest of the industry, is set up to do. What they’re incentivized to do — and it made sense when we started — was simply to invest more capital. But that doesn’t make sense anymore, and it’s not really helping us.
Now, that’s not intended to vilify the utility industry — although I will say it’s a very target-rich environment if anyone wants to go do that. It’s simply an observation of the fundamental economics and physics. I’ll offer a slight diversion: in my younger years I worked as an EMT, and we’re taught some basic things about ABCs, starting with airway. If the airway isn’t open, nothing else matters. If we don’t face this challenge about how the utility industry is set up right now — so that we’re driving toward the wrong thing — we don’t have the airway open to let these technologies and solutions breathe. You can do all the CPR you want; it’s not going to work if the airway isn’t open. So this is just something we need to address, and I think that conversation is being forced, from data centers down to the other end of the spectrum, where people are plugging things in at their home or business and they don’t need permission. So it’s happening at all scales. That’s what I’d say about scale.
Greg Castle [15:30]
The only thing I’d add is that I also think it’s a very unique opportunity in our industry — with the amount of capital being deployed for data center infrastructure — to look at it slightly differently, as opposed to the negative connotation right now. How do you flip it and drive more community benefits with the investment, to allow them to connect faster?
We’re seeing some innovative examples, like in New Jersey, where they’re saying you can interconnect faster if you offset the peak grid constraint by funding and deploying individual batteries in residential communities or residential homes. So if there’s a two-hundred-megawatt grid constraint at that transmission feeder, you could offset that with a virtualized VPP that you aggregate. It would drive down affordability, increase reliability and resilience, and increase sustainability. It’s a win-win-win situation. There are some very creative approaches being explored and implemented — it’s more a question of how we take those examples and rapidly copy-paste them to scale faster.
Mike Grenier [16:40]
Can I add one thing? I want to reframe the question a little bit, because I think this is happening whether or not — it’s just going to happen. It’s a snowball rolling downhill at this point. So it’s not a question of “can we scale.” We actually have everything we need to scale. It’s going to happen, I promise. The question is maybe how we accelerate it. And I think both of you make great points about the blockers.
In there was a theme of a clear price signal. If the end consumer has a clear price signal that using power in the afternoon is expensive, they will respond to that. If the consumer has a clear price signal that installing a battery will get them a capacity payment, they will respond to that. So the utilities can really play a huge role in accelerating the deployment in a way that’s really exciting.
Elisa Wood [17:40]
Great points, thank you so much. One thing we’re sometimes missing is that energy is in a unique position right now. Data centers have a real problem — they have a community-opposition problem — and energy has the solutions. It’s amazing when you think about the things data centers can bring to communities by offering the right energy, whether it’s solar and storage through a virtual power plant, combined heat and power providing heat, or microgrids and the resilience they’ll bring. So they have a big problem, and energy has the solution. It’s a very interesting position to be in these days.
Newsletter break
Elisa Wood [18:30]
Let me stop to share something. I’ve noticed, after years covering electric power, that the real story of how the grid is changing hardly ever shows up in a single headline. It’s happening piece by piece — in microgrids, solar, virtual power plants, and new market models that reshape how electricity works. That’s what I track in the Energy Changemakers newsletter. Each week, I break down the most important developments in distributed energy and explain what they mean for the industry and the world. Recent issues have looked at the rise of data center microgrids, controversies over virtual power plants, and why balcony solar shouldn’t be ignored. If you enjoy this podcast and want a clear weekly briefing on where the power system is heading, you can subscribe for free at energychangemakers.com. Now, let’s get back to the conversation.
Audience question: equity and who benefits
Elisa Wood [19:20]
I want to stop for a second and see if anybody has a question they’d like to ask this great panel. Yes?
Audience member [19:30]
The question is, will it scale, or how will it scale? I’d like to make sure that we’re thinking about the way it scales in the community. We’re in a place where it’s really hard for people to buy homes. So how do we ensure that when we’re scaling this, we’re not giving all of the money to the people who are already benefiting?
Elisa Wood [19:55]
Good question — how do we ensure that the right people get the benefits? Cameron.
Cameron Brooks [20:05]
I think it’s an excellent question — really the key question that almost everything should be centered around. That should be front and center from the beginning. To me, one of the things that’s both exciting and, perhaps, terrifying — but that gets to your question — is that it comes down to ownership. Right now, we more or less have a model where we’re all renting our energy.
If you think about it — and I don’t want to get too lost in this, but we’re just coming out of the Fourth of July — the American Dream really comes down to ownership, and to community building. We’re renting our energy because, under most of the rules right now, the only people allowed to invest are the utility industry. And yet we’re seeing all this activity around that system we’ve set up, where companies are investing, consumers are investing directly, and more and more, communities are too.
So part of the way we ensure the benefits go where they should is by bringing it down to the local level. That’s one of the things I find really interesting and exciting about where we are: the technology is finally there for us to have all of these local solutions. It’s getting cheaper to produce an electron and cheaper to store an electron. Really, the only thing getting more and more expensive is to move that electron around on the legacy electric grid. So I think we’ll see more and more benefit when we begin to relax that a little bit.
We’re already seeing some of that in certain communities that are creating — and I know this term has legal definitions — essentially safe harbors. Here in Oregon there was legislation to create microgrid zones. I haven’t tracked the proceeding play-by-play, and I’m sure a lot of people in this room have more direct experience with it, but that’s a model many communities are looking at. So I think that’s at least one way to get at the idea of how you make sure the benefits stay with the people: through ownership. And ownership means both legal ownership and some accountability and voice in the process — and I think that happens at the local level.
Audience question: interconnection and siting barriers
Elisa Wood [22:30]
Okay, any other questions? We’ll give you another chance later, too, but we’ve got one over here.
Jake Edie (RenewComm) [22:40]
Hi, Jake Edie from RenewComm. When you think about large-scale energy development, the two biggest problems or challenges right now are interconnection and community acceptance. Those are mainly absent from on-site generation, or much, much smaller. When you think about the scale problem we were just talking about, how important is the absence of those two blockers in driving that scale?
Producer note: The audience member’s name and firm (”Jake Edie from RenewComm”) are transcribed here as our best-verified reading of the original audio, which rendered the firm as “Renucalm.” Please confirm against the recording if the spelling matters for publication.
Mike Grenier [23:00]
I’ll speak from experience, having done both large generation and distributed. Large-generation project timelines could easily be six, seven, or ten years, and a good chunk of that is what you mentioned. Just last quarter, we installed a new solar and battery system at a building — I think it was done in three months. So that’s the difference.
Greg Castle [23:30]
From the interconnection standpoint, I think the majority of these off-grid builds are because of the long interconnect process. The data centers need to come online as fast as possible and don’t have the luxury of waiting three, five, or seven years for an interconnection. I don’t want to speak on behalf of data centers, but my guess is the majority would grid-connect if they had the option within the timeframe they need. So I think that’s a major blocker, and a major push to go off-grid, because of the speed-to-power requirements.
Cameron Brooks [24:00]
I don’t know that I have a solution to the interconnection challenge, but I think it’s a great question, because it is a challenge. One of the things we’re seeing — there are articles out there that talk about how data centers, for example, are about to all go off-grid and become this phantom load operating independent of the system. Personally, I think that’s a possibility, and it’s certainly happening for all the reasons we’ve talked about; economically, it just makes sense for people to do things at all scales, from the data center down to the home.
But it would be one of the worst outcomes if that ended up with a completely fragmented system. So I would love for us to think about how we don’t look at interconnection as a blocker — although it is right now — but come up with ways to invite companies, communities, businesses, and consumers to interconnect, to be interactive, and to bring the load flexibility we’ve talked about that data centers might have. This is an opportunity.
Mike, you said it: this is going to happen. People are investing in this stuff, and we can’t stop them. Again, here in the West — if the power is being shut off — my power was turned off in December for five days, right around the new year, and that’s an unacceptable situation. I’m glad they turned the power off so they didn’t spark a wildfire; that would be worse. But it’s an unacceptable solution. So there’s going to come a point where people say, “You can’t tell me I can’t have this stuff that makes economic sense and brings resilience.” They’re going to do it. The question is how we harness that — going back to ownership and community — so that it brings benefits to all of us.
I do come back to this moment of great optimism, because almost anything is possible right now. Data centers are changing the situation, technology is changing the situation, and the increasing vulnerability we face from weather events is changing the situation. I’m pretty excited about what the next two or three years might look like. In the face of all of this, which sounds like huge problems — and they are challenges — I’m actually pretty optimistic that we’re going to come out with some really good solutions.
Elisa Wood [27:00]
I agree completely, Cameron, that this is a good time. We start to get caught up on the problems, because we’re all working in the field and have to deal with these obstructions. But wow, this is an incredible time. So I want to get back to our regular programming and talk about the economics of data centers. Mike, I want to start with you, because you made a point that data centers aren’t just creating demand — they’re creating the economic signal that finally tips the teeter-totter. Explain to us how that works economically.
The economics of data centers
Mike Grenier [27:20]
Sure. I think data centers are really shining a spotlight on the capacity problem the grid has. We have a shortage of capacity, and when — what did you say, 115 gigawatts? — 115 gigawatts, France and the UK, show up all of a sudden wanting to plug into the grid, how are you going to get that quickly?
What the data centers can do that is unique from lots of small loads connecting to the grid is that they come with scale. So you can see interesting deals being cut, like the one between Google and Voltus, where Google is actually going to pay to put distributed batteries across the grid in order to solve that capacity problem for the grid, so they can come online faster — because it’s faster to do that than to go build a big central station. That’s the power the data centers have to really accelerate the transformation.
Greg Castle [28:30]
To double down on that: on average, the current utility grid utilization in the U.S. is around 52%. So 52% of the grid, on average, is being used, and the grid is obviously built for those one or two peak days, either in the summer or the winter depending on geography. So the opportunity in the short term, if we could solve for this flexibility challenge and figure out that secret sauce at scale, is huge — there’s a huge amount of capacity left if you can solve for the capacity during those one or two peak-constraint days. We’re talking four to eight hours in a year that need to be solved for.
Cameron Brooks [29:20]
It’s interesting. I think there are two constraints. There’s a capacity constraint: these data centers are connecting and they want power, the grid can’t deliver it, and the grid can’t be built out fast enough. And there’s the challenge around peak load, where at the same time that we don’t have enough power, most of the time most of the grid is sitting around idle. So it’s a funny paradox.
What data centers are bringing is exactly like what you described — data center companies partnering with others that can provide flexibility, resilience, and some amount of capacity that the grid needs. So again, it’s a question of making sure we don’t brand those companies and data centers as defectors, put some kind of scarlet letter on them — maybe two letters, “DC” — and treat them as a pariah. We want to bring them in and say, “This is actually what we need.”
Right now in Virginia, the State Corporation Commission is trying to figure out how to actually incentivize the flexibility we need. It’s not an easy problem to solve, but that’s one of the big opportunities right now. From my point of view, having worked on this for a long time and being very excited about distributed energy, the challenges an individual home faces are exactly the same as what a data center faces from a regulatory and interconnection point of view.
It’s not just the physical interconnection that’s important — there’s a data and information interconnection that goes along with it. When should I be dispatching? How should we respond to grid conditions? And there’s an economic interconnection: am I going to get access to markets and be able to receive the value I’m bringing? So we don’t just want to focus on the physical interconnection; these same challenges exist all up and down the scale.
What data centers are bringing that hasn’t existed before is this: an individual homeowner can’t really go knock on the door and say, “Hey, I want you to treat me fairly.” But suddenly here come huge dollars, huge demand, and a lot of ability to shine a spotlight on a problem that’s existed for a long time, and to try to come up with solutions. So I think we’re going to see more and more of partnerships like the one you brought up.
When you get down to the fundamentals, we’re looking for data centers to be flexible. And one of the things that maybe isn’t discussed — maybe it’s a dirty little secret — is that data centers probably aren’t flexible. They have the incentive to bring flexibility to the system, but the data center itself is probably not a flexible load. Some maybe are, but overall what they can bring is through partnerships: how do we put batteries out in the community, and work with the community to bring what they need, without expecting data centers to fundamentally change their operations — any more than we’d ask a car manufacturer or any other industry to fundamentally change how they operate to serve the grid? But I think there are a lot of solutions.
The “Frankenstein microgrid”
Elisa Wood [32:40]
Thank you. So, our panelists don’t know this, but there was a secret vote at Energy Changemakers. We had a pre-call where we talked about what we were going to discuss, and I let some of the folks at Energy Changemakers listen to it. They decided we had to give out an award to the person who came up with the best new term for the industry. That award is going to Greg Castle of Schneider Electric, and his new term is a “Frankenstein microgrid.” So Greg, explain to us what a Frankenstein microgrid is.
Greg Castle [33:20]
Let me recall that. This goes back to the energy management system, the protection controls, and the overall software layer, and how critical it’s become. A lot of these projects were very COVID-like — go out and buy whatever you can — and we’ve seen this with a lot of these large projects. It was kind of a gold rush when this thing hit twelve or eighteen months ago. Everyone went out and secured whatever they could: aeroderivatives, refurbished airplane engines, pretty much every recip engine that could be bought, every turbine. You can’t get a turbine until 2033 right now.
So they showed up with hundreds of megawatts, if not gigawatts, of these generating assets, and they’re trying to figure out how to put it together — not only put it together and start it up as fast as possible, but then optimize this kind of Frankenstein-ed, multi-gigawatt worth of generating assets, optimizing it to both the utility and the data center itself. It’s extremely complex. It’s not like they intentionally went out and did it; they just bought what they could, and now they’re trying to put it together. I think we’re going to see a lot of these projects get put together over the next couple of years. It’s a very solvable problem, but the moral of the story is: don’t underestimate the protection controls and the energy management system needed to really optimize that solution and to mitigate the O&M expenses on the back end. So, I don’t know if that explained it.
Audience question: advice for students
Elisa Wood [35:20]
That was great. We can take maybe one more question. Yes.
Kat (Clackamas Community College) [35:30]
My name is Kat. I work with Clackamas Community College now, and I’m a former student of our renewable energy program. As a key takeaway, what would you say are some of the most valuable insights you’d like to give to former, current, and aspiring students who are going into energy — working with utilities and interconnections?
Elisa Wood [35:55]
Great — she’s asking for some key points, a great wrap-up.
Mike Grenier [36:00]
Sure — what should you make sure to focus on when you’re in school? We’re in the middle of doing a lot of hiring right now, so this is front of mind. We’re really pulling in all kinds of skills; a lot of the things applicable in other industries are applicable here. But the basics are still the most important. Somebody who knows how to learn fast, somebody who knows how to be a clear and effective communicator — those skills are not getting less valuable, they’re getting more valuable. AI is making them more valuable. So I would not neglect that part. Everything else is learning the nuances of an industry and the technical details, but that foundation is still critical.
Greg Castle [36:50]
I don’t know if this will completely answer it, but I’ll just say: what an exciting time to be getting into the industry. The energy industry probably hasn’t been sexy for a while, but now, all of a sudden, it’s very exciting. And to Skip’s earlier comment — what’s ironic is that these AI data centers are causing a lot of the current constraints, but at the same time, AI is also what’s going to solve a lot of these problems when we’re talking about large-scale grid scenarios. The potential with AI is just getting started.
From our perspective, the hardware aspect is obviously part of it, but the software really has unlimited potential, and it’s just getting started. There are so many startups; it’s such an exciting time to get into the industry. And not to mention, everything’s starting to shift to DC. Thomas Edison is rolling over — now we’re architecting 800-volt-DC data centers, and we’re deploying DC microgrids, because the storage, the solar, the cars — everything’s DC. And the power electronics — medium-voltage solid-state transformers, the heavy power electronics — are just getting developed now and will be deployed in the coming years. So it’s super exciting to be starting out in this industry.
Cameron Brooks [38:20]
I would agree — it’s a great time. In some respects, I’m jealous of you to be standing here at the beginning of it. I have three answers for you. The first: I tend to look at things through certain lenses, but I think policy is really important. We underestimate how much words on paper and laws that come out have an impact on the physical landscape, the physical grid. You can’t learn that in a textbook — or at least I haven’t seen that textbook. Most people learn it by getting out in the field, and one of the good or bad things about that is that to get out in the field, you kind of have to put on a jersey. So you’re playing for one team or another, and there are different teams battling against each other. I would not underestimate how much influence and importance that has.
Having said that, I would also say: follow the physics. The physics and the thermodynamics are hard to argue with. There are a lot of reasons why it makes sense for power to be delivered locally, and for the solution that makes the most sense to be put in place. There are probably a whole bunch of tangents we could go down related to thermodynamics — we tend to look at things through the first law, and maybe we should be looking more at the second law of thermodynamics, which would guide us to more local solutions.
The last thing I’d say, which builds off what you offered, is to look at history a little bit, and at some other industries. There’s often an analogy made, at least in regulatory circles, around telecom — what happened there, how that industry changed, and what that might mean for the regulation of the utility industry. One thing that’s often not talked about enough is that one of the things that changed how telecom works was decoupling the physical part of the network. We don’t have switchboards anymore, where someone needed to plug in to make a physical connection and keep it open so you could have a conversation. We turned it into digital packets that could be broken up, sent, stored, and moved along.
I think the analogy is very apt here, because right now we have this synchronous grid that needs to be managed all the time — and yet we have an opportunity, especially with storage, to follow the exact same path: to decouple that and move energy around in packets, not on a synchronous grid that needs to be maintained in every moment. So those are my thoughts. Thank you for the question — I think it’s great, and keep at it.
Wrap-up
Elisa Wood [41:00]
Okay, so that wraps it up. I want to thank this amazing panel: Cameron Brooks from Think Microgrid, Greg Castle from Schneider Electric, and Mike Grenier from Elexity. This was a wonderful panel — thank you so much for being here. A lot of great information, and a great audience, thank you. And thank you, Robert, and thank you, Skip, for putting on this great event. Looking forward to the next couple of days.
Announcer [41:40]
Thank you for joining us today on the Energy Changemakers podcast. I hope today’s conversation has inspired you to think differently about the role of distributed energy in our society. If you enjoyed this episode, please subscribe to our podcast on your favorite platform and consider leaving us a review. You can help us shape the dialogue around energy. Go to energychangemakers.com, subscribe to our newsletter, and join our community. Together, we can make a real difference in how quickly we can make our grid local, equitable, and clean.
